Full text of "Proceedings of the Linnean Society of New South Wales"

See other formats

PROCEEDINGS
of the

LINNEAN

SOCIETY

NEW SOUTH WALES
VOLUME 136

NATURAL HISTORY IN ALL ITS BRANCHES

THE LINNEAN SOCIETY OF
NEW SOUTH WALES
ISSN 1839-7263

Founded 1874
Incorporated 1884

The society exists to promote the cultivation and
study of the science of natural history in all branches.
The Society awards research grants each year in the
fields of Life Sciences (the Joyce Vickery fund) and
Earth Sciences (the Betty Mayne fund), offers
annually a Linnean Macleay Fellowship for research,
and publishes the Proceedings. It holds field
excursions and scientific meetings including the
biennial Sir William Macleay Memorial Lecture
delivered by a person eminent in some branch of
natural science.

Membership enquiries should be addressed in the first instance to the Secretary. Candidates for
election to the Society must be recommended by two members. The present annual membership
fee is $45 per annum.

Papers are published at http://escholarship.library.usyd.edu.au/journals/index.php/LIN and access
is free of charge. All papers published in a calendar year comprise a volume. Annual volumes are
available to any institution on CD free of charge. Please notify the Secretary to receive one. “Print
on demand” hardcopies are available from eScholarship.

Back issues from Volume 1 are available free of charge at www.biodiversitylibrary.org/title/6525.

OFFICERS and COUNCIL 2014

President: R.J. King

Vice-presidents: M. Gray, D. Keith, D.R. Murray
Treasurer: I.G. Percival
Secretary: J-C. Herremans

Council: M.L. Augee, J.P. Barkas, M. Cotton, E. Gorrod, M.R. Gray, J-C. Herremans, D. Keith,

R.J. King, H.A. Martin, D.R. Murray, P.J. Myerscough, I.G. Percival, J. Pickett, H.M. Smith, B.

Welch and K.L. Wilson

Editor: M.L. Augee

Assistant Editor: B. Welch

Auditors : Phil Williams Carbonara

The postal address of the Society is P.O. Box 82, Kingsford, N.S.W. 2032, Australia.

Telephone and Fax +61 2 9662 6196.

Email: linnsoc@iinet.net.au

Home page: www.linneansocietynsw.org.au/

Cover motif: Palaeoxyris duni holotype MMF 42697a (left) and paratype MMF 42697b (right)
on a single slab shown in Figure 2 in the paper by Graham McLean, pages 201-218, this volume.

PROCEEDINGS
of the

LINNEAN

SOCIETY

NEW SOUTH WALES

For information about the Linnean Society of New South Wales, its publications and

activities, see the Society’s homepage
www.linne ans o cietyns w. org.au

VOLUME 136

December 2014

Mi n erals of Jenolan Caves, New South Wales, Australia:
Geological and Biological Interactions

R. E. Pogson 1 , R. A. L. Osborne 1 ’ 2 , and D. M. Colchester 1

Geoscience and Archaeology, Australian Museum, 6 College St, Sydney NSW 2010 (ross.pogson@austmus.
gov.au); 2 Faculty of Education and Social Work, University of Sydney, NSW 2006

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Pogson, R.E., Osborne, R.A.L. and Colchester, D M. (2014). Minerals of Jenolan Caves, New South
Wales, Australia: geological and biological interactions. Proceedings of the Linnean Society of
New South Wales 136 , 1-18.

Geological and biological processes in the Jenolan Caves have formed a range of mineral species
spanning several chemical groups. So far 25 mineral species have been either confirmed, or identified for
the first time at Jenolan. Their chemical groups include carbonates: (calcite, aragonite, hydromagnesite,
huntite, dolomite, ankerite); silicates: (kaolinite, K-dehcient muscovite (‘illite’), montmorillonite clays);
phosphates, (ardealite, hydroxylapatite, taranakite, leucophosphite, variscite, crandallite, montgomeryite,
kingsmountite); sulfate: (gypsum); oxides: (quartz, cristobalite, amorphous silica, hematite, romanechite);
hydroxide: (goethite); nitrate: (niter); and chloride: (sylvite). Dolomitised limestone bedrock and ankerite
veins can be recognised as a magnesium source of some magnesium carbonate minerals, as well as
supplying a calcite inhibitor favouring aragonite formation. The cave clays have diverse origins. Some
are recent sedimentary detritus. Older clays of Carboniferous age contain components of reworked altered
volcaniclastics washed or blown into the caves, so these clays may represent argillic alteration of volcanic
products. Some of the clays may have formed as alteration products of ascending hydrothermal fluids. The
phosphates and some gypsum formed when bat guano reacted chemically with limestone and cave clays.
Gypsum has also been formed from the breakdown of pyrite in altered bedrock or dolomitic palaeokarst.
The niter and sylvite have crystalized from breakdown products of mainly wallaby guano.

Manuscript received 9 August 2013, accepted for publication 23 April 2014.

KEYWORDS: biology, geology, guano, Jenolan Caves, minerals

INTRODUCTION

The Jenolan Caves, 182 km west of Sydney in
the Greater Blue Mountains World Heritage Area, are
developed in the folded, steeply dipping, Late Silurian
Jenolan Caves Limestone (Came and Jones 1919;
Chalker 1971; Allan 1986), and in places intersect
stratified Palaeozoic marine carbonate palaeokarst
deposits, termed caymanite (Osborne 1991, 1993,
1994). The Jenolan Caves Limestone extends 7 km in
a NW direction, in the valley of McKeown’s Creek,
having a maximum thickness of 265 m. The caves are
developed on the northern and southern sides of the
Grand Arch, a natural tunnel cutting the limestone.

From the discovery of Jenolan Caves by
Europeans after 1838, the calcite speleothems have
been admired for their beauty and variety, but other
minerals were not recognised until the late 1 9 th Century,

when niter was first reported by Wilkinson (1886),
and gypsum and phosphate minerals by Mingaye
(1898). An Australian Museum-Sydney University
project to characterise the minerals of Jenolan Caves
has been active for over 25 years, and so far has either
confirmed, or identified for the first time at Jenolan,
25 mineral species. Some of the minerals have been
formed by completely inorganic chemical processes,
but others have involved chemical interactions with
biological materials such as marsupial and bat guano.
Mineral species in the nitrate, chloride, carbonate,
phosphate, sulfate, silicate, oxide and hydroxide
chemical groups are described.

MATERIALS AND METHODS

The studied mineral specimens, collected by
the authors over the last 25 years, are lodged in the
Australian Museum Geoscience collections. Earlier

MINERALS OF JENOLAN CAVES

cave mineral specimens were donated to the Australian
Museum by J.C. Wiburd in 1898 and by others
in the late 19 th and early 20 th centuries. Specimen
registration numbers prefixed by ‘D’ or ‘DR’ are in
the Australian Museum Mineralogy and Petrology
collections respectively, and other designations are
specimen field collection numbers.

The minerals were identified using a variety of
methods, including field observations and analytical
methods. Most of the identifications were carried out
at the Australian Museum using X-ray diffraction
(XRD) equipment (PANalytical X’Pert Pro) with a
graphite monochromator, proportional counter, and
45 kV, 40 mA Cu-Ka radiation. Scans were run from
5-70° 20 with 1° divergence slit, 2° antiscatter slit,
0.1 mm receiver slit and 0.02° steps, with additional
scans 2-15° 20 with 0.125° divergence slit and 0.25°
antiscatter slit. Peak patterns were processed using
PANalytical X’Pert HighScore software.

X-ray fluorescence analysis was carried out at
the University of Technology, Sydney, using an Rh
target at 60 kV, 40 mA, LiF 420, LiF 220, Ge 111
and T1AP analysing crystals, and Siemens UniQuant
software. Energy dispersive X-ray spectrometry
(EDS) was carried out at the Australian Museum, with
an Oxford Instruments Link Isis 200 EDS coupled
with a Cambridge Stereoscan 120 SEM, with internal
Co standard, in backscatter electron mode. Spectra
were accumulated for 100 seconds at 20 kV, 18-25
x magnification, and 20 microsecond processing
time. SEM imaging was performed at the Australian
Museum with a Leo 435VP and later a Zeiss EVO LS
15, using gold coated samples mounted on standard
SEM stubs, at 15-20 kV and 18-23 mm working
distance.

Laser Raman spectroscopy was performed at
the Queensland University of Technology using a
Renishaw 1 000 Raman system, with a monochromator,
filter system, CCD detector (1024 pixels) and
Olympus BHSM microscope equipped with lOx,
20x, and 50x objectives. The spectra were excited by
a Spectra-Physics model 127 He-Ne laser producing
highly polarised light at 633 nm and collected at a
nominal resolution of 2 cm 1 and precision of ± 1 cm 1
in the range between 200 and 4000 cm' 1 . Repeated
acquisitions using a 50 x microscope objective were
accumulated to improve the signal to noise ratio of
the spectra. Raman Spectra were calibrated using
the 520.5 cm -1 line of a silicon wafer. The spectra
of at least 10 crystals were collected to ensure the
consistency of the data. Infrared spectra were collected
at the Queensland University of Technology using a
Nicolet Nexus 870 FTIR spectrometer with a smart
endurance single bounce diamond ATR cell. Spectra

over the 4000 - 525 cm -1 range were collected with a
resolution of 4 cm -1 and a mirror velocity of 0.6329
cm s' 1 . Spectra (128 scans) were co-added to improve
the signal to noise ratio.

Other methods used were polarized light
microscopy, ultraviolet fluorescence, K-Ar dating
(CSIRO Petroleum) and fission track dating (Geotrack
International), inductively-coupled plasma mass
spectrometry (University of Cape Town), and sulfur
and oxygen isotope determinations (Environmental
Isotopes, Sydney, and University of Barcelona).

RESULTS

The following species catalogue documents
the rich mineralogical diversity of Jenolan Caves,
including physical descriptions and mode of
occurrence for those minerals known up to July 2013.
The species and their formulae are listed in Table 1 .

Calcite

Calcite is the trigonal form of calcium carbonate.
It is the main mineral in the Jenolan Caves Limestone,
which has an average composition of 97.6% CaC0 3
(Sussmilch and Stone 1915; Came and Jones 1919;
Chalker 1971; Allan 1986). It also contains small and
variable amounts of Mg and Fe. It forms the majority
of cave speleothems in an incredible variety of forms,
including stalactites, stalagmites, columns, straws,
shawls, shields, canopies, helictites, cave pearls, rim
pools, pool crystal, rafts, cave coral and flowstone.
The spectacular calcite speleothems are the main
features of Jenolan Caves, and a selection is shown
in Fig. l:a,b,c,d,e. Calcite is also deposited through
interaction between colonies of blue-green algae
(cyanobacteria) and drip water, forming rounded,
crenulated stalagmites (stromatolites or ‘cray backs’)
found in Nettle Cave (Fig. 2) and the Devil’s Coach
House. A speleothem classification is presented in
Hill and Forti (1997).

Calcite also forms curious fluffy growths of
tangled microcrystalline filaments. They form light
insubstantial masses of ‘fairy floss’ or cotton wool
appearance (Fig. 3a) and have been observed up to 5
-6 cm diameter. When damp the masses are coherent,
but fall apart to a white powder when dry. SEM images
of these fluffy growths from Wilkinson Branch of
Chifley Cave show microcrystalline aggregates of
tangled filaments 0.5-1 micron diameter (Fig. 3b). It
often grows on a porous mud or clay substrate, but can
also cover rock, flowstone and stalactites. It is mainly
calcite, with minor silica and water. They have been
aptly described by (Mingaye 1899:330):

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Table 1: Identified Jenolan Caves mineral species

Chemical formulae are from the International Mineralogical Association

Commission on New Minerals, Nomenclature and Classification

approved list. The ‘illite series’ chemical formula is from Rieder et al.

( 1998 ).

ankerite

aragonite

ardealite

calcite

crandallite

cristobalite

dolomite

goethite

gypsum

hematite

huntite

hydromagnesite

hydroxylapatite

kaolinite

kingsmountite

leucophosphite

montgomeryite

montmorillonite

muscovite, K-deficient

niter

quartz

romanechite

sylvite

taranakite

variscite

Ca(Fe,Mg,Mn)(C0 3 ) 0

CaC0 3

Ca„(P0 3 0H)(S0 4 ).4H,0

CaC0 3

CaAl 3 (P0 4 ) 0 (OH) 5 .H 0 0

SiO,

Ca(Mg,Fe)(C0 3 X

Fe 3+ 0(0H)

CaS0,.2H,0

Fe 2 0 3

CaMg 3 (C0 3 ) 4

Mg 5 (C0 3 ) 4 (0H) ? .4H 0 0

Ca 5 (P0 4 ) 3 (0H)

Al ? Si 2 0 5 (0H) 4

(Ca,Mn 2+ ) 4 (Fe 2+ ,Mn 2+ )Al 4 (P0 4 ) 6 (0H) 4 .12H,0
KF e 3 T(PO X(OH) 2H,0

occurs on limestone cave walls,
palaeokarst deposits and cave
clays.

Dolomite

Veins of iron-bearing

dolomite intersect the Jenolan
Caves Limestone and one bed
towards the top of the Limestone
is extensively dolomitised.

An Fe-rich dolomite mass
can be seen just in front of
the entrance to the Lyrebird’s
Chamber, Ribbon Cave. Near
the western edge of the Jenolan
Caves Limestone in Contact
Cave, dolomitic stalactites are
currently forming. Dolomitic
palaeokarst (caymanite) occurs
extensively in the Mud Tunnels
section of River Cave.

Ca 4 MgAl 4 (P0 4 ) 6 (0H) 4 .12H,0
(Na,Ca) 03 (Al,Mg) 2 Si 4 O 10 (OH) 2 .nH 2 O
(“illite series”) K 065 A1 20 dA1 065 Si 335 O 10 (OH) 2

kno 3

Si0 2

(Ba,H,O) o (Mn 4+ ,Mn 3+ ) s O 10

KC1

K 3 A1 5 (P0 3 0H) 6 (P0 4 G 1 8H,0
A1P0 4 .2H,0

“Two samples of this substance were received. The
first, which weighed E4 grammes, Guide Wiburd
states, was compressed into a small match-box, and
would fill your hat in its natural state. It is so light
that, when you blow on it, it falls off the roof and
sides like snow”.

Aragonite

Aragonite, the orthorhombic form of calcium
carbonate is the second most common cave mineral
world-wide (Hill and Forti 1986, 1997), although its
occurrence is still relatively rare. Aragonite crystal
groups are highly regarded for their aesthetic value
and can form some of the most spectacular of all
speleothems (Fig. 4).

At Jenolan, aragonite fonns white stalactites,
straws, columns, helictites, needles, ‘flos ferri’
and anthodites (quill-like crystal sprays). It is
found in a number of caves, including Ribbon (the
Lyrebird’s Nest), Pool of Cerberus (the Arabesque),
River (the Furze Bushes), Jubilee, Red, Chevalier,
Wiburd ’s Lake, Mammoth, Spider, Glass, Contact
and Barralong Caves (Rowling 2004, 2005a, b). It

Ankerite

Ankerite forms thin yellow
to brown branching veins up
to several tens of centimetres
length in the limestone (Fig. 5),
and can also form larger zones
of replacement. Ankerite also
mantles some of the fossils
in the limestone. Much of the
veining has a surface alteration
to goethite and clays, sometimes
with minor silica. The unaltered material has a sugary
texture due to small rhombohedral crystals of ankerite
up to 0.3 mm.

Huntite and hydromagnesite

These two minerals are the major components of
a Mg-carbonate-rich ‘moonmilk’, a white, structure-
less, plastic, spongy mass with high water content,
sometimes with a ‘cauliflower’ appearance, and
having the consistency of cottage cheese when damp,
but falling apart as a white powder when dry. Huntite
and hydromagnesite with minor calcite, aragonite, and
silica have been observed as a 4 cm mass growing on
an aragonite stalactite in the Lyrebird’s Nest, Ribbon
Cave. Under SEM, this moonmilk shows rosettes of
platy crystals up to about 5 microns (Fig. 6). Traces
of huntite and hydromagnesite have been found in
Wiburd’s Lake Cave (Rowling 2005b).

Clays (Kaolinite, K-deficient muscovite (“illite”),
montmorillonite)

Cave clays of plastic consistency and white,
yellow, brown and red colours are widespread. Most

Proc. Linn. Soc. N.S.W., 136, 2014

Figure la,b,c,d,e:
Calcite speleothems
a: Minaret, River

MINERALS OF JENOLAN CAVES

• r* r

« -3
u 1
S3

<u
£

•- s

a> 5

I — ;> w .

.VuU I
| 3 ’oS S w

® ^ T

^ © c8 -+5 -S3 .2
S3 53 U S £ 73 S3

•c ^ 5 ta & ® -C

r * L,

5® ,_>

5 ® 5 ®

• • rs 3

g S3 s

it © §

5 ® §5

3 M 3

5<£S

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Figure 2: Stromatolites (‘craybacks’), Nettle Cave, width approx. 1.5 m. Image: Ross Pogson, Austral-
ian Museum

are kao 1 inite/il 1 ite mixtures, often with quartz, and
sometimes with minor calcite or montmorillonite,
and coloured by hematite or goethite.

As well as massive forms, kaolinite can also
form pseudo-hexagonal crystals up to 3 microns (Fig.
7a), and also pseudomorphs tiny feldspar crystals of
possible volcanic origin, in pink clay (DCH4) from
the Devils Coach House. Illite forms delicate fluffy or
hairy growths on kaolinite (Fig. 7b).

Ardealite

Rounded yellow bosses up to about 9 cm high,
called ‘point 068 ’ by Foster (1890); Trickett (1905);
and Havard ( 1 928), form the “Potato Patch” of Dunlop
(1979), in the Bone Cave section of Lucas Cave (Fig.
8a, b). Similar deposits occur in the Grotto Cave
section of Chifley Cave (Fig. 8c), and were reported
by Mingaye (1898, 1899). He analysed but did not
name the ardealite, identified the accompanying
gypsum, and noted that similar deposits were seen
in the Bone Cave, Lucas Cave. The ‘potatoes’ have
a thin outer shell of gypsum, but their interiors are
filled with softer cream-coloured powdery ardealite,

and minor calcite. They sit on a layer of mixed
ardealite, gypsum and hydroxylapatite. Although they
sit on a sloping surface, they display a vertical growth
axis and may have been deposited under subaerial
conditions by vertically-drawn solutions (Pogson
et al. 2011). Ardealite is uncommon in Australia,
but Bridge (1967) and Bridge et al. (1975) reported
ardealite with brushite from Marooba Cave, Jurien
Bay, Western Australia, and Grimes (1978) identified
ardealite from Texas Caves, Queensland, all with
guano associations.

Hydroxylapatite

Hydroxylapatite is found in a number of places as
thin white coatings and small nodules. In the Grotto
Cave chamber of Chifley Cave, hydroxylapatite has
formed under ardealite ‘potatoes’, and occurs with
crandallite. It is part of a complex phosphate mineral
mixture in Katie’s Bower, Chifley Cave. In Lucas
Cave, traces are present on walls near The Slide; in
layers below the ‘potatoes’; lining a solution tube in
limestone above the ‘potatoes’; and with crandallite
in Bone Cave. It has also been found on cave earth

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Figure 3a, b: a: Calcite fluffy growth, Wilkinson Branch of Chifley Cave, width 6 cm. Image: Ross Pog-
son; b: SEM image, calcite fluffy growth, Wilkinson Branch, Chifley Cave. Field of view 141 x 98 mi-
crons. Image: Sue Lindsay, Australian Museum

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Figure 4: Aragonite, Chevalier Cave. Image: Ted Matthews, Jenolan Caves

Figure 5: Ankerite veins in limestone, 30 m south of eastern entrance to Grand Arch. View approx. 20 x
15 cm. Image: Ross Pogson, Australian Museum

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Figure 6: SEM image, Mg-rich moonmilk, Lyrebird’s Nest Chamber, Ribbon Cave. Field of view 50 x 35
microns. Image: Sue Lindsay, Australian Museum

coating limestone in Queen Esther’s Chamber,
River Cave. The CO, -rich variety (‘carbonate-
hydroxy lapatite’) occurs in Katie’s Bower, Chifley
Cave, as amber botryoidal crusts of 0.5 mm
spherules. Hydroxylapatite is present in altered bat
guano collected by Wiburd in 1898 from Lucinda
Cavern, Chifley Cave (DR12132) and in altered
guano collected more recently from the Exhibition
Chamber of Lucas Cave. It is also intermixed with
other phosphate minerals elsewhere.

Crandallite

Crandallite fonns a white, chalky, crumbly
deposit in the Bone Cave section of Lucas Cave,
collected by guides from floor deposits (D58074).
In Chifley Cave it occurs as a white crumbly nodular
deposit (D56949) (Fig. 9) on red-brown clay on
limestone, above the ‘potatoes’ alcove in the Grotto
Cave (Pogson et al. 2011), and as a component of
phosphates in Katie’s Bower.

Other phosphates - Katie’s Bower, Chifley Cave
(Taranakite, Variscite, Montgomeryite,
Kingsmountite, Leucophosphite)

These five phosphate minerals, together
with hydroxylapatite, C0 2 -rich hydroxylapatite,
crandallite, calcite, silica, and horizontally- stratified
clays (kaolinite, illite) have been identified in Katie’s
Bower, Chifley Cave, as complex intergrowths in a
prominent poorly-consolidated, unstable outcrop.
This deposit has slumped downwards and forward,
obscuring contacts and extending under the path and
down a gentle slope to the opposite cave wall. Many
of the phosphates fluoresce pale yellow under short-
wave ultraviolet light.

Mingaye ( 1 898) reported and analysed phosphates
from this location, including taranakite (then called
‘minervite’) in ‘Left and Right Imperial Caves’.
As to their origin, Wiburd found mineralised bat
guano in Lucinda Cavern, Chifley Cave (his donated
specimen DR12132). Taranakite, hydroxylapatite,
leucophosphite and variscite (D58072) form crumbly,
fine-grained white to cream-coloured chalky
masses and irregular veins and coatings. An iron-
bearing variety of variscite is also present (KB 19).
A taranakite specimen (misidentified at the time as
‘sulphate of alumina’) (D 10948) was found in ‘New
Cave’ (probably Jubilee Cave) by Robert Etheridge,

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Figure 7a, b: SEM images, a: Kaolinite crystals (JRV7), River Lethe, River Cave, field of view 20 x 14
microns; b: Illite on kaolinite (JRV9), The Junction, River Cave, field of view 48 x 34 microns. Images:
Sue Lindsay, Australian Museum

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Figure 8a,b,c: a: ‘Potatoes’, Bone Cave, Lucas Cave; b: Ardealite ‘potato’, Lucas Cave (D49535), size 60
mm;, c: ‘Potatoes’, Chifley Cave (D57257), size 70 mm. Images: Ross Pogson, Australian Museum

Australian Museum in 1896. Taranakite is uncommon
in Australia, but Bridge (1967) reported it from caves
near Jurien Bay, and on the Nambung River, Western
Australia.

Montgomeryite is the Mg-analogue of
kingsmountite (Anthony et al. 2000). Montgomeryite

forms tiny rosettes of pale yellow platy crystals 0.05
-0.3 mm and micro-botryoidal aggregates similar to
those of kingsmountite, in small crystal-lined cavities
in white chalky matrix, and as a minor component
of massive, crumbly, chalky mixed phosphates.
Additional tests by Frost et al. (2012a) confirmed the
identification as montgomeryite (D58702, D58703).

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Kingsmountite forms white radiating thinly
bladed to acicular microcrystals with silky lustre,
in compact aggregates of spherules 2-3 mm. It
also forms larger bladed microcrystals in crusts up
to a 4 mm thick (D58703). Blocky fawn-coloured
crystals less than 0.5 up to 1 mm line small vughs
in fine-grained crumbly, chalky matrix. It is closely
associated with montgomeryite. SEM images show
the kingsmountite rosettes are made of 2 - 3 micron
crystal plates (Fig. 10).

Gypsum

Gypsum forms the yellow outer shell of the
‘potatoes’ in Lucas Cave (Bone Cave section) and in
Chifley Cave (Grotto Cave section). Mingaye (1898,
1 899) analysed the Grotto Cave deposit and identified
the material as gypsum. It is also found as white to
colourless deposits elsewhere in small cavities in
upper Bone Cave, and crusts in Centenary Cave, in
Lucas Cave. Minor deposits also occur in other caves.
In the past, gypsum has been found as white to clear
masses and curved, fibrous crystal groups (gypsum
‘flowers’) associated with palaeokarst (D 19994) in
the Devil’s Coach House. Their original collection site
in the Devil’s Coach House is unknown but gypsum
specimens from this locality are preserved in the
Australian Museum collections (D 19994, D 12021)
(Fig. lla,b).

Silica (quartz, cristobalite, amorphous)

Silica is present as small quartz grains in many
clays and cave sediments. It also occurs as minor
poorly crystalline cristobalite and amorphous silica
in mixed phosphates; in fluffy calcite growths in the
Wilkinson Branch of Chifley Cave; and in Mg-rich
moonmilk in Ribbon Cave.

Romanechite

Romanechite is present as thin black to brown-
black surface coatings in many places. It is also
present as tiny rounded black nodules to 1.5 mm in
clays.

Goethite and hematite

Goethite and hematite are widespread. They
are present in red, brown and yellow clays, and also
stain calcite speleothems in a variety of colours.
Extensive concretionary goethite deposits, often as
‘pipe concretions’ are seen near the far end of Jubilee
Cave.

Niter and sylvite

Niter was reported by Wilkinson (1886) from
the Grand Arch, and by Mingaye (1898, 1899) from

the Devil’s Coach House. More recently, Sydney
University researchers (J. James pers. comm.) found
niter and tiny cubic crystals of sylvite in dust from
the SW side of the Grand Arch. Jenolan Caves staff
found thick crusts of white material, later identified
as niter, from Dust Cave high up on the south side of
Grand Arch. Dust Cave isa9x7x2m cavity just
to the right of the top of the Lucas Cave path, partly
in between large limestone boulders fallen from the
roof. The cave walls consist of limestone boulders
and bedrock, conglomerate cave fill and lithified
palaeokarst.

The niter crusts in Dust Cave are horizontal
deposits on the cave floor, partly covered by a
large pile of fine grey-brown dust. Small crusts and
stalactites of niter occur on the cave walls, especially
on porous conglomeratic cave fill. These niter crusts
are up to 5 cm thick (D52263) and have a corroded
appearance, best seen in cross-section (Fig. 12).
The corrosion channels run perpendicular to the
horizontal crust surface. The crusts contain skeletal
crystals of sylvite 0.1-3 mm (average 0.3 mm),
sometimes occurring in long strings perpendicular
to the horizontal surface of the crusts. The sylvite
content of the crusts is variable, ranging from 5-15 %
by volume (Colchester et al. 2001). Crevice fillings of
niter occur in several places in the Grand Arch.

DISCUSSION

Carbonates

The calcite speleothems are usually very pure
calcium carbonate but can be stained various colours
by iron oxides and hydroxides, manganese oxides
and organic material. Calcite can also contain
magnesium, manganese, strontium and iron in its
crystal structure.

Formation of the calcite-rich fluffy growths is
still a matter for debate. Suggestions have been: it
formed as a part of the life cycle of bacteria, fungi
or algae; it is a disintegration product of bedrock or
other cave formations; or it is a precipitation directly
from ground water, but special conditions promote
fibrous crystal aggregates. It has also been suggested
the flat calcite lamellae are pseudomorphs after
metastable monohydrocalcite (hexagonal CaC0 3 .
H 2 0), or that the calcite needles are paramorphs after
aragonite (Onac and Ghergari 1993; Ghergari et al.
1994). These growths also occur in Chevalier, Glass,
Mammoth, and Wiburd’s Lake Caves. A chemical
analysis of fluffy calcite growths from Wilkinson
Branch, Chifley Cave, is given in Table 2 (analysis
3).

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Figure 9: Crandallite, Grotto Cave, Chifley Cave. Size 10 x 25 cm. Image: Ross Pogson, Australian
Museum

Figure 10: SEM image, Kingsmountite (D58703), Katie’s Bower, Chifley Cave. Field of view 280 x 190
microns. Image: Sue Lindsay, Australian Museum

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Figure 11a, b: a: Gypsum, Devils Coach House, D12021, 14 x 7 x 5 cm (with attached palaeokarst); b:
Gypsum ‘flowers’ to 80 mm (D19994). Images: Ross Pogson, Australian Museum

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Fig. 12: Niter/sylvite crust (D52263), Dust Cave, Grand Arch.
Size 7x5x5 cm. Image: Stuart Humphreys, Australian Mu-
seum

Aragonite is normally stable at higher
temperatures and pressures over 3 kbar (MacDonald
1956). However, it forms in caves at ambient
pressures and temperatures, (approximately 1 5°C and
1 bar), which is outside its thermodynamic stability
field. Based on P/T data alone, only calcite should
be present, and aragonite will very slowly alter to
calcite but seems to persist for a long time in caves.
It is widely thought that aragonite is formed when
calcite precipitation is inhibited by structural poisons
like Sr, Fe, Mg, phosphate and sulfate ions, but other
factors, such as low drip rates, low evaporation rates,
and variations in temperature, atmospheric humidity
and carbon dioxide concentration may be involved. If
the presence of Mg ions is a factor in its formation,
possible Mg sources are nearby dolomite or ankerite.
Jenolan aragonite has approximately 0.3% SrO as
well as minor Mg, Fe, P etc. (Table 2, analyses 1 and
2 ).

The ankerite and dolomite have variable
chemistry. They are more susceptible than limestone
to alteration, and it is possible that much of the
soggy goethite-bearing clay deposits in caves along
McKeown’s Valley, near the western limestone
contacts were formerly iron-bearing dolomites. Much
of the horizontally-bedded Carboniferous carbonate
palaeokarst (caymanite) in The Mud Tunnels,

River Cave is dolomitic. Some of this
palaeokarst contained pyrite (now
altered to goethite), and square crystal
outlines can be seen in thin-section.
Compositions of Fe-rich dolomite from
near the entrance of the Lyrebird’s
Nest Chamber, Ribbon Cave; and from
Contact Cave near the western edge
of the Jenolan Caves Limestone, are
shown in Table 2, analyses 4 and 5.

Mg-rich ‘moonmilk’, a mixture of
microcrystalline magnesium-bearing
carbonates, is often found in close
proximity to aragonite, or dolomite
and ankerite. The dolomite is often
decomposed to a goethite or hematite-
rich clay, obscuring its carbonate origins .
The decomposition of these Mg-rich
minerals provides the sources of Mg 2+
ions in solution required for formation
of huntite and hydromagnesite.
These minerals are precipitated when
magnesium ion concentration increases
with evaporation (Hill and Forti 1997).

Clays

The clay deposits appear to represent different
ages and origins. Younger clays may originate from
wind-blown dust (loess), or water-borne sediments
brought in by streams or floodwaters. However,
some of the older clays appear to have components
of altered, reworked volcanic ash detritus washed
or blown into the caves. This volcanic detritus is
present in River Lethe clay, River Cave. Some of the
clay deposits may have formed as alteration products
resulting from ascending hydrothermal fluids. The
evidence for presence of upwelling hydrothermal
fluid activity is based on the analysis of cupola
morphology by Osborne (1999). Large deposits of
water-bearing goethite-rich clays occurring in caves
along McKeown’s Valley, near the western edge of
the Jenolan Caves Limestone may represent altered
dolomitised limestone bedrock.

The potassium content of K-deficient muscovite
(‘illite’) made Potassium- Argon dating possible, and
a range of illite-bearing clays gave Carboniferous
dates of 320-357 Ma (mean 337 Ma) (Visean to
Namurian). Details of the K-Ar dating results are
presented in Osborne et al. (2006, Table 5). Fission
track dating of zircon grains extracted from clay
from The Junction, River Cave, gave a central age
of 308.9+/-25.6 Ma with two age groups with pooled
ages of 435.9+/- 19.1 Ma (Carboniferous) and 207.2
+/- 18.5 Ma (Late Triassic to Early Jurassic) (Green

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Table 2: Analyses of some Jenolan Caves aragonite, calcite and dolomite (XRF wt%)

1. Aragonite (JRV2) from a stalactite near the Furze Bush chamber, Mud Tunnels, River Cave.

2. Aragonite (JR7A) from crystal spheres, Lyrebird’s Nest Chamber, Ribbon Cave.

3. Calcite fluffy growth (Wl), from Wilkinson Branch, Chifley Cave.

4. Dolomite, iron-bearing (J167), near entrance of Lyrebird’s Chamber, Ribbon Cave.

5. Dolomite, iron-bearing (JC3), from a stalactite in Contact Cave.

Analyst: Marie Anast, University of Technology, Sydney, by X-ray fluorescence (total Fe as Fe 2 Q 3 )

SiO,

0.14

7.1

H0 2

0.35

ai 2 o 3

0.07

3.5

MnO

0.46

Fe A

0.08

3.29

MgO

0.86

k 2 o

0.02

1.4

CaO

99.3

82.6

BaO

0.02

0.03

SrO

0.26

0.28

0.15

so 3

0.04

Total

99.9

100.0

2003). Details of the fission track dating results are
presented in Osborne et al. (2006, Table 6). It is
suggested in Osborne et al. (2006) that some of these
Carboniferous clays and their associated sand-size
fraction (pyroclastic rock fragments, pyroxenes,
zircons and illite pseudomorphs after feldspar) have
a common origin, likely derived from reworked local
Carboniferous volcaniclastics. The pristine crystal
forms of both kaolinite and illite in several Jenolan
clay deposits indicate that they have been allowed to
grow in situ, undisturbed (Figs 7a, b).

Phosphates

Pogson et al. (2011) suggested that the ardealite
‘potatoes’ in Lucas and Chifley Caves have grown
vertically from a sloping surface, under subaerial
conditions, growing upwards from their porous
substrate base by evaporation of vertically-drawn
pore water (wicking). Their morphology and mode of
occurrence make formation in a pool or by dripping
water less likely. Sulfur isotopes in ardealite gave
5 34 S of +11.12 to +12.8 °/ 00 (Pogson et al. 2011, Table
3), suggesting that sulfate-rich solutions came from
leaching of bat guano deposits (at Jenolan from the
bent-wing bat, Miniopterns schreibersii ). These

0.79

5.8

11.7

0.02

0.37

0.11

0.35

3.64

5.1

0.02

0.87

0.07

0.45

6.8

2.47

0.23

9.2

5.7

0.11

1.62

1.58

97.5

71.3

72.4

0.27

0.03

0.06

0.35

0.29

0.24

0.09

0.10

0.04

99.9

100.0

99.9

mineralised solutions reacted with limestone (Pogson
et al. 2011). Most cave phosphate minerals originate
from guano, which can also be a source of sulfur
(Hill and Forti 2004). There is currently no visible
bat guano near the ‘potatoes’ in Lucas Cave, although
old deposits do occur in the Exhibition Chamber in
another part of the Lucas Cave system. Ardealite
chemical analyses are presented in Pogson et al. (20 1 1 ,
Table 2). Identification was confirmed by additional
analytical methods (Frost et al. 2011a, 2012b).

Hydroxylapatite was probably a precursor
mineral in the formation of other phosphates from
leached bat guano (Marincea et al. 2004), being
altered by changes in ion concentrations, Eh, pH and
temperature of the percolating solutions. Crandallite
is often an earlier-formed phosphate. Additional
confirmatory tests for this Jenolan crandallite are
detailed in Frost et al. (2011b, 2012c).

The complex aluminium, calcium and potassium-
bearing phosphate minerals of Katie’s Bower, Chifley
Cave formed from reaction of acidic phosphatic
solutions leached from bat guano, with limestone
and cave clays. The phosphate deposits are complex
mixtures of fine-grained minerals, making XRD
identification difficult. Although bat guano was

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

found in the 19 th Century in the Lucinda Cavern of
Chifley Cave, no recognisable guano traces remain in
Katie’s Bower. This suggests the mineral assemblage
represents the final stage of guano alteration,
consistent with pH levels approaching neutral (Vince
et al. 1993). The chemical reactions have also released
amorphous silica.

Variscite can be fonned by leaching of
montgomeryite with the loss of calcium (Hill and
Forti 1997). Kingsmountite (Dunn et al. 1979) is a
rare mineral. It is even rarer in cave environments,
being previously reported only from Rossillo Cave,
Mexico (Forti et al. 2006), although it occurs in USA,
Russia, Portugal, Germany, and South Australia in
other types of phosphate deposits (Anthony et al.
2000 ).

Onac and Veres (2003) and Marincea et al (2004)
discussed formation of phosphates in Romanian
caves, finding that ardealite could fonn from pre-
existing hydroxylapatite. In general, the phosphates
were formed from phosphate-rich, then sulfate-rich
solutions, accompanied by pH changes reflecting
the degree of carbonate dissolution. Hydroxylapatite
usually fonns at higher pH, and is destabilised for
pH values up to 5.5, but if sulfur is available under
those conditions, ardealite formation is favoured.
Taranakite forms early under damp conditions in the
presence of excess potassium, from solutions with
pH < 6. Vince et al. (1993) studied the paragenesis of
phosphate minerals at the Parwan Cave in Victoria,
Australia, finding that taranakite and clay fonned
early, with apatite later, and finally montgomeryite.
An additional confirmatory test for Jenolan taranakite
is detailed in Frost et al. (2011c).

Sulfates

Gypsum was considered the second most common
cave mineral by White (1976) and Onac (2005), and
the third most common after calcite and aragonite
by Hill and Forti (1997). Gypsum sulfur isotope
signatures from Lucas and Chifley Cave ‘potatoes’
gave a S 34 S of +1 1 .3 to +1 1 .8 °/ 00 indicating an organic
origin, derived from bat guano (possibly via sulfur-
oxidising bacteria in the guano). Sulfur isotopes from
gypsum in the Devil’s Coach House deposits (S 34 S of
+1 .4 to +4.9 °/ 00 ) indicate an inorganic origin, probably
from breakdown of pyrite in carbonate palaeokarst.
This interpretation of the isotope data is discussed in
Pogson et al. (2011, Table 2).

Oxides and hydroxides

Apart from quartz sand grains, silica is widespread
as a minor microcrystalline or cryptocrystalline
component of many cave minerals, and it has also

been released by chemical alteration of clays by acidic
phosphatic solutions, giving rise to poorly crystalline
cristobalite, as well as amorphous silica. Hematite
and goethite in the caves may have originated from
a variety of processes, including the breakdown of
iron-bearing carbonates and pyrite, ferruginous wind-
blown dust, and deposition from percolating iron-rich
solutions. The high barium content of the clays (0.05
- 0.096 wt%) is due to the presence of romanechite,
which occurs as small rounded concretionary nodules,
and is also common as thin films and coatings
elsewhere (Osborne et al. 2006).

Nitrates and chlorides

The dust pile in Dust Cave is leached, powdery
dung from the brush-tailed rock wallaby Petrogale
penicillata. Waters percolating through the dung
have leached nitrates and chlorides and redeposited
them as niter and sylvite. The Grand Arch is very dry,
and in winter cold, dry westerly winds blow straight
through, promoting evaporation and deposition of the
niter and sylvite crusts. The sylvite was deposited
after the niter, filling some of the vertical cavities
between the niter crystals (Colchester et al. 2001).

ACKNOWLEDGEMENTS

The authors are grateful to the Jenolan Caves Reserve
Trust, the Karst and Geodiversity Unit of National Parks
and Wildlife Service, and the Manager, Jenolan Caves,
for providing access to the Caves and for permission for
the mineral sampling. The Jenolan Caves guides gave
invaluable assistance and advice at all times. This study
included specimens collected under provisions of Scientific
Licences S12664 and SL100197 of NSW National Parks
and Wildlife Service, for Jenolan Karst Conservation
Reserve. Sue Lindsay, Manager, SEM Unit, Australian
Museum, provided the SEM images. The authors thank
both Reviewers for their helpful comments, which have
clarified and improved the manuscript.

REFERENCES

Allan, T.E. (1986). Geology of Jenolan Caves Reserve.

BSc (Hons) Thesis, University of Sydney (unpubl.).
Anthony, J.W., Bideaux, R.A., Bladh K.W. and Nichols
M.C. (2000). "Handbook of Mineralogy- Vol. IV
(Arsenates, Phosphates, Vanadates)’ . (Mineral Data
Publishing: Tucson, Arizona).

Bridge, P. (1967). Mineral, Mineral Technology and
Geochemistry Division. In ‘Report of the Western
Australian Government Chemical Laboratories
for the Year 1966’, 42-43. (Western Australian
Government: Perth).

Proc. Linn. Soc. N.S.W., 136, 2014

R.E POGSON, R.A.L. OSBORNE AND D.M. COLCHESTER

Bridge, P., Hodge, L.C., Marsh, N.L., Thomas, A.G.

(1975). Chiropterite deposits in Marooba Cave,

Jurien Bay, Western Australia. Helictite 13, 19-34.

Carne, J.E. and Jones, L. J. (1919). The Limestone

Deposits of New South Wales. Mineral Resources of
New South Wales , 25, 1-383.

Chalker, L.E. (1971). Limestone in the Jenolan Caves
area. Records of the Geological Survey of New South
Wales. 13(2), 53-60.

Colchester, D.M., Pogson, R.E. and Osborne, R.A.L. (2001).
Niter and sylvite from Jenolan Caves, New South
Wales, Australia. Proceedings of the 13 th International
Congress of Speleology, Brasilia D.F., Brazil, July
15-22, 2001. Session (SI) Geospeleology. 133-SI, 4p.
Published on CD.

Dunlop, B.T. (1979). ‘Jenolan Caves’. (New South Wales
Department of Tourism: Sydney). 11 th edition.

Dunn, P.J., Pecor D.R., White, J.S. and Ramik, R.A.

(1979). Kingsmountite, a new mineral isostructural
with montgomeryite. Canadian Mineralogist 17, 579-
582.

Forti, P, Galli, E. and Rossi, A. (2006). Peculiar

minerogenetic cave environments of Mexico: the
Cuatro Cienegas area. Acta Carsologica 35(1), 79-98.
(Ljubljana, Slovenia).

Foster, J. J. (1890). ‘The Jenolan Caves’. (New South
Wales Government Printer: Sydney).

Frost, R.L., Palmer, S., Henry, D A. and Pogson, R.

(2011a). A Raman spectroscopic study of the ‘cave’
mineral ardealite Ca 0 (HPO 4 )(SO 4 ).4H 2 O. Journal of
Raman Spectroscopy 42(6), 1447-1454.

Frost, R.L., Xi, Y, Palmer, S.J. and Pogson, R. (2011b).
Vibrational spectroscopic analysis of the mineral
crandallite CaAl 3 (P0 4 ) 2 (0H)..(H 2 0) from the Jenolan
Caves, Australia. Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy 82, 461-
466.

Frost, R.L., Xi, Y, Palmer, S.J. and Pogson, R. (2011c).
Vibrational spectroscopic analysis of taranakite
(K,NH 4 )A1 3 (P0 4 ) 3 (0H).9H 2 0 from the Jenolan Caves,
Australia. Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy 83( 1 ), 1 06- 111.

Frost, R.L., Xi, Y, Palmer, S.J. and Pogson, R. (2012a).
Identification of montgomeryite mineral [Ca 4 MgA
1 4 (P0 4 ) 6 (0H).12H 2 0] found in the Jenolan Caves,
Australia Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy 94, 1-5.

Frost, R.L., Palmer, S.J. and Pogson, R.E. (2012b).

Thermal Stability of the ‘cave’ mineral ardealite
Ca 2 (HP0 4 )(S0 4 ).4H,0. Journal of Thermal Analysis
and Calorimetry 107(2), 549553.

Frost, R.L., Palmer, S.J. and Pogson, R.E. (2012c). Thermal
Stability of crandallite CaAl 3 (P0 4 ) 2 (0H) 5 .H,0: a ‘cave’
mineral from the Jenolan Caves. Journal of Thermal
Analysis and Calorimetry 107(3), 905-909.

Ghergari, L., Onac, B.P and Santamarian, A. (1994).
Mineralogy of moonmilk formation in some
Romanian and Norwegian caves. Theoretical and
Applied Karstology 6, 107-120.

Green, PF. (2003). ‘Fission Track Dating of a zircon
concentrate from Jenolan Caves. Geotrack Report #
872’. Uunpublished report to the Australian Museum,
March 2003. (Geotrack International Ltd: Brunswick
West, Victoria, Australia).

Grimes, K.G. (1978). The geology and geomorphology of
the Texas Caves, Southeastern Queensland. Memoirs
of the Queensland Museum 19 (1), 17-59.

Havard, W.L. (1928). ‘The Lucas Cave, Jenolan Caves:
a practical guide to the regulation Route’. (New
South Wales Government Tourist Bureau: Sydney,
Australia). Facsimile edition (Jenolan Caves
Historical and Preservation Society, 2011).

Hill, C.A. and Forti, P. (1986). ‘Cave Minerals of the
World’, (National Speleological Society: Huntsville,
Alabama USA).

Hill, C.A. and Forti, P. (1997). ‘Cave Minerals of the
World’, Second Edition. (National Speleological
Society: Huntsville, Alabama USA).

Hill, C.A. and Forti, P. (2004). Minerals in Caves. In
‘Encyclopedia of Caves and Karst Science’ (Ed. J.
Gunn), 511-514.

MacDonald, G.J.F. (1956). Experimental determination
of calcite-aragonite equilibrium relations at elevated
temperatures and pressures. American Mineralogist
41(9-10), 744-756.

Marincea, A., Dumitras, D-G., Diaconu, G. and Bilal, E.
(2004). Hydroxylapatite, brushite and ardealite in the
bat guano deposit from Pestera Mare de la Meresti,
Persani Mountains, Romania. - Neues jahrbuch fur
mineralogie- monatshefte 10, 464-488.

Mingaye, J.H. (1898). On the occurrence of phosphate
deposits in Jenolan Caves, New South Wales.
Australasian Association for the Advancement of
Science, Report 5, 423-425

Mingaye, J.H. (1899). On the occurrence of phosphate
deposits in Jenolan Caves, New South Wales.

Records of the Geological Survey of New South Wales
6(2), 11-116.

Onac, B.P. and Ghergari, L. (1993). Moonmilk mineralogy
in some Romanian and Norwegian caves. Cave
Science 20(3), 107-111.

Onac, B.P. (2005). Minerals. In ‘Encyclopedia of Caves’
(Eds. D. Culver and W.B. White), 371-378.

Onac, B.P. and Veres, D.A. (2003). Sequence of secondary
phosphates deposition in a karst environment:
evidence from Magurici Cave (Romania). European
Journal of Mineralogy 15, 741-745.

Osborne, R.A.L. (1991). Palaeokarst deposits at Jenolan
Caves, New South Wales. Journal and Proceedings
of the Royal Society of New South Wales 123, 59- 73.

Osborne, R.A.L. (1993). Cave formation by exhumation
of Palaeozoic palaeokarst deposits at Jenolan Caves,
New South Wales. Australian Journal of Earth
Sciences 40, 591-593.

Proc. Linn. Soc. N.S.W., 136, 2014

MINERALS OF JENOLAN CAVES

Osborne, R.A.L. (1994). Caves, dolomite, pyrite, aragonite
& gypsum: The karst legacy of the Sydney and
Tasmania Basins. In ‘Twenty-Eighth Newcastle
Symposium on Advances in the Study of the Sydney
Basin’, University of Newcastle Department of
Geology Publication 606 , 322-324.

Osborne, R.A.L. (1999). The origin of Jenolan Caves:
elements of a new synthesis and framework
chronology. Presidential Address for 1998 - 1999.
Proceedings of the Linnean Society of New South
Wales 121 , 1-27.

Osborne, R.A.L., Zwingmann, H., Pogson, R.E. and
Colchester, D.M.. (2006). Carboniferous clay
deposits from Jenolan Caves, New South Wales:
implications for timing of speleogenesis and regional
geology. Australian Journal of Earth Sciences 53 ( 3 ),
377-405.

Pogson, R.E., Osborne, R.A.L., Colchester, D.M.
and Cendon, D.I. (2011). Sulfate and phosphate
speleothems at Jenolan Caves, New South Wales,
Australia. Acta Carsologica 40 ( 2 ), 239-254.
(Ljubljana, Slovenia).

Rieder, M., Cavazzini, G., D’Yakonov, Y.S., Frank-
Kamenetskii, V.A., Gottardi, G., Guggenheim, S.,
Koval, P.V., Muller, G., Neiva, A.M.R., Radoslovich,
E.W., Robert, J.-L., Sassi, L.P, Takeda, H., Weiss, Z.,
and Wones, D .R. (1998). Nomenclature of the micas.
The Canadian Mineralogist 36 , 905-912.

Rowling, J. (2004). Cave aragonites of NSW. Master
of Science Thesis (unpubl.). (The School of
Geosciences, Division of Geology and Geophysics,
The University of Sydney).

Rowling, J. (2005a). Studies on aragonite and its

occurrence in Caves, including New South Wales
Caves. Journal and Proceedings of the Royal Society
of New South Wales 137 ( 3 - 4 ), 123-149.

Rowling, J. (2005b). Cave aragonite in NSW, Australia.
Proceedings of the International Union of Speleology
Congress, Greece, Session 18, Mineralogy in Caves
II, Talk 0-88.

Sussmilch, C.A. and Stone, W.G. (1915). Geology of the
Jenolan Caves district. Journaland Proceedings of the
Royal Society of New South Wales 49 ( 3 ), 144-148.

Trickett, O. (1905). ‘Guide to the Jenolan Caves, New
South Wales’. (New South Wales Government
Printer: Sydney).

Vince, D., Hall, P. and Birch, W. (1993). Phosphate
minerals in cave deposits. In ‘Phosphate Minerals
of Victoria’ ( Eds W.D. Birch and D.A. Henry) 121-
151. (The Mineralogical Society of Victoria Inc.:
Melbourne)

White, W.B. (1976). Cave minerals and speleothems. In
‘The Science of Speleology’ (Eds T.D. Ford and
C.H.D. Cullingford) 267-327.

Wilkinson, C.S. (1886). ‘Railway Guide to New South
Wales’. (New South Wales Government Printer:
Sydney).

Proc. Linn. Soc. N.S.W., 136, 2014

The Jenolan Environmental Monitoring Program

Andrew C. Baker

Karst and Geodiversity Unit
National Parks and Wildlife Service
Office of Environment and Heritage
Level 2, 203-209 Russell Street, Bathurst, NSW 2795

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Baker, A.W. (2014). The Jenolan environmental monitoring program. Proceedings of the Linnean Society
of New South Wales 136 , 19-34.

The Jenolan Environmental Monitoring Program reports on the condition of atmospheric and water
parameters in and around the show caves at Jenolan. This paper summarises the key findings from four
years (2009-2012) of monitoring cave atmosphere. The caves were typically characterised by high relative
humidity, moderately stable air temperature (annual variation <2°C) and pronounced seasonal variation in
the concentration of C0 2 . A major exception was the Temple of Baal, where C0 2 was moderately elevated
(-2,000 ppm) year round, with no apparent seasonal variation. The concentrations of C0 2 in the caves were
generally well below the exposure limit and pose minimal risk to human health.

Abrupt increases in air temperature of up to 0.9°C in <12 minutes occurred in several of the caves, in
particular the Imperial. These increases were characteristic of, and generally corresponded to, commercial
tours, and rapidly stabilised back to the pre-tour temperature after the tour had passed. Similarly, increases
in C0 2 associated with visitation were generally short lived, except in the Temple of Baal, where peak
visitation elevated the CO, for extended periods of time. The merits and shortfalls of various options for
managing the accumulation of CO, in the Temple of Baal are discussed.

Manuscript received 5 December 2013, accepted for publication 16 April 2014.

KEYWORDS: carbon dioxide, cave atmosphere; Jenolan Caves, temperature, tourist cave.

INTRODUCTION

The maintenance of natural processes within karst
environments is highly dependant on the interactions
between the soil, water and air (Watson et al. 1997).
In recognition of these interactions, karst managers
around the world are increasingly utilising air and
water monitoring programs to examine the impacts of
humans in show caves (e.g. Pulido-Bosch et al. 1997;
Russell and MacLean 2008; Lario and Soler 2010).
Indeed Cigna (2004) recommends that the atmosphere
in all show caves should be monitored to determine
whether cave operations are having adverse impacts
on these unique and delicate environments.

Jenolan Caves are arguably Australia’s best
known show caves, attracting more than 200,000
visitors each year (Jenolan Caves Reserve Trust
2012). In 2004 the Jenolan Caves Reserve Trust
(JCRT) received $4.2 million to undertake capital
works within Jenolan Karst Conservation Reserve
(JKCR). Of this funding, approximately $200,000

was allocated to the development of an air and
water quality monitoring program. Sites for air and
water monitoring and equipment were selected in
2006 in consultation with the Jenolan Scientific and
Environmental Advisory Committee, with monitoring
equipment purchased in 2007 and progressively
installed in 2008-2009.

The Jenolan Environmental Monitoring Program
(JEMP) measures air and water quality parameters
that are of relevance to karst conservation, the
maintenance of biological diversity and visitor safety.
The measurement and reporting of such parameters
enables an objective evaluation of the environmental
performance of JCRT (or any future successor), with
regards to air and water quality in the show caves.
Specifically, the JEMP aims to achieve this by
establishing and reporting trends in air quality and the
relationship between these trends and anthropogenic
activity (in particular commercial cave tours) and
trends in water quality at a number of sites with the
catchment of the tourist caves. This paper presents

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

the key findings from four years (2009-2012) of
monitoring cave atmosphere in the show caves at
Jenolan.

METHODS

Sites

The Jenolan Karst Conservation Reserve
(JKCR), is located approximately 1 80 km southwest
of Sydney (33°49’S, 150° 02’E) and one of eight
reserves that comprise the Greater Blue Mountains
World Heritage Area (DECC 2009). Monitoring of
cave atmosphere was conducted at seven sites within
the show caves and one external reference site. Three
monitoring sites were situated in the Northern Show
Caves (Chifley, Diamond and Imperial) and four
sites in the Southern Show Caves (Mafeking Branch
of Lucas, Orient, River and Temple of Baal (Fig. 1,
Table 1).

Equipment

Commercially available instruments manu-
factured by Vaisala were utilised to measure
air temperature (Vaisala ‘HMT 100’, ± 0.2°C),
relative humidity (Vaisala ‘HMT 100’, ±2.5% RH),
barometric pressure (Vaisala ‘PTB 110’ ±0.3 hPa) and
the concentration of carbon dioxide (C0 2 ) (Vaisala
‘GM220’ ±1.5% of range ± 2 % of reading) at each
monitoring site. Monitoring equipment was housed
within a sealed case at each site and connected to
the cave power supply. A data logger (ACR Systems
Inc. ‘SmartReader Plus 7’) recorded each of the
parameters every 6 minutes. This interval was chosen
following preliminary trials in 2008, which found
6 minute intervals were a sufficient frequency to
capture the influence of passing tour groups, whilst
providing sufficient storage time (2 months 12 days)
before the logger began to rewrite over the oldest data.
Data were downloaded every 2 months for analysis
and inclusion in bi-monthly Condition Reports that
are provided to the JCRT by the OEH Karst and
Geodiversity Unit.

Data analysis

Data were filtered using the ‘macro’ and ‘IF’
functions in Microsoft Excel to remove data that
exceeded a maximum permissible temperature range
between the temperature probe and the data logger
(<8°C) or maximum permissible change between
two consecutive time internals (<2°C change in
six minutes). These values were determined from
the results of preliminary trails in 2008, which
ascertained the largest genuine difference between
the temperature probe and the data logger (i.e. the

□ Jubilee

□ Imperial/Diamonci

□ Elder

□ Chifley

■ Nettle

H Devils Coach House

Grand Arch

Lucas/Lurline

River/Pool of Cerberus

Temple of Baal

Orient

Ribbon

Figure 1. Map of the tourist cave system at Jeno-
lan. Monitoring sites are numbered as per the de-
scription in Table 1. Image courtesy of the Jenolan
Survey Project.

Proc. Linn. Soc. N.S.W., 136, 2014

A. BAKER

Table 1. Air monitoring sites

Name

Site number

Site description

Northern Show Caves

Chifley Cave

In Katie’s Bower, on the tourist platform above the mains power
distribution

Diamond Cave

Approximately 5 m beyond the end of the public viewing area.

Imperial Cave

Nellie’s Grotto

Southern Show Caves

Lucas Cave

In the Mafeking branch approximately ten steps below the
highest point of elevation

Orient Cave

In Lower Indian Chamber

River Cave

Under the bridge approximately 15m before the Pool of
Reflections

Temple of Baal

At the middle junction next to the switchboard

External site

School House

On the south facing wall.

temperature in the sealed case), and largest change
between consecutive temperature readings. Filtering
the data was necessary to remove false values that
occurred as a result of power outages and surges.
Where temperature values at a site were false, it
was assumed that all other parameters at that point
in time were also false. The final step of filtering the
data involved manually checking the graphed data for
anomalies. For example, it was common for several
sites to simultaneously experience an abrupt decrease
of all parameters and such values were regarded as
false and removed from the dataset.

Bimonthly datasheets were combined into a single
spread sheet to determine the average, maximum and
minimum values for each of the parameters for the
four year period. The daily average air temperature,
relative humidity and C0 2 were calculated for each
day from January 1, 2009 to December 31, 2012 and
graphed to examine any variation in cave atmosphere
between seasons from one year to the next. Lastly,
to ascertain the influence of commercial cave tours
on the cave atmosphere, air temperature, relative
humidity and CO, data, one week of continuous data
(i.e. readings taken every six minutes) for April 5-11
2012 was graphed at a larger scale. This period was
chosen as it included Easter (April 6-8), typically one
of the busiest periods of visitation during the year.

RESULTS

Overview of cave atmosphere

The average air temperature in the caves ranged
from 11.7°C in the Diamond Cave to 15.5°C in the
Temple of Baal (Table 2, Fig. 2a). In comparison to
the external site, where the average air temperature
was 12°C and ranged from -4°C in winter to >
30°C summer, the air temperature within the caves
was highly stable. The Temple of Baal had the least
variation, with a temperature range of 0.6°C, while
the Chifley experienced the largest range (3.2°C,
Fig. 2a). Similarly for relative humidity (RH), while
the external site experienced a highly degree of
variability (6.9 - 99.6 % RH), the cave atmosphere
was characterised by very high (98.8 - 99.9 %) and
stable RH (Table 2, Fig. 2b). As with air temperature,
the Chifley experienced the largest range in RH
(9.7 %). However given the low variability of RH
compared to the precision of the probes (± 2.5%),
detailed analysis was not possible.

Whereas the caves experienced considerably less
variation in air temperature and RH than the external
site, this was not the case for C0 2 , with the cave
atmosphere recording a much larger range in CO,
than the external atmosphere (Fig 2c). In comparison
to the external atmosphere (-380 ppm) the average
concentration of CO, at the monitoring sites within

Proc. Linn. Soc. N.S.W., 136, 2014

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

Table 2. Average data (2009-2012) for air temperature, relative humidity and the concentration of C0 2
at each of the monitoring sites.

Chifley

Diamond

Imperial

Lucas

Orient

River

Temple of
Baal

External

Temperature (°C)

Mean

12.98

12.65

13.04

14.24

14.94

13.36

15.49

12.01

0.33

0.09

0.11

0.06

0.11

0.20

0.05

7.13

Range

3.22

1.02

1.57

2.03

0.94

1.66

0.63

Relative humidity (%)

Mean

98.75

99.74

99.85

99.69

99.12

99.61

99.47

72.63

0.59

0.17

0.10

0.01

0.31

0.06

18.61

Range

9.72

4.37

3.91

2.73

3.79

4.48

1.89

92.74

C0 2 (ppm)

Mean

1,013.4

848.0

857.2

1,098.8

759.2

718.6

2,142.1

381.4

853.1

275.6

413.5

461.3

347.1

333.4

425.3

35.2

Range

4,847.9

1,596.6

2,333.3

2,022.2

2,292.3

1,668.4

2,690.6

352.1

16 1

15 -

—

14 -

13 -

12 -

11 -

10 -

T I

Z 1

. T

i 1

1 1 1 1 1 1

Chifley Diamond Imperial Lucas

Orient

River

Temple of
Baal

Site

Figure 2. Average a) air temperature, b) relative humidity (RH) and c) C0 2 concentration at each of the
monitoring sites from January 1 2009 to December 31 2012. The high-low lines show the range (maxi-
mum and minimum values). Temperature and RH data for the external site are not shown due to the
high degree of variability, [b) and c) on following page].

Proc. Linn. Soc. N.S.W., 136, 2014

A. BAKER

Figure 2 continued

l»

■a>

■ HH

Baal

Site

6000
| 5000
^ 4000
* 3000

”

= 2000
.-■-i

c 1000
cj

i i i j

1 1 1

1 i

1 1 1 1

Chifley Diamond Imperial Lucas Orient

River Temple of External
Baal

the caves ranged from 848 ppm in the Diamond to
2,142 ppm in the Temple of Baal. While the average
concentration of C0 2 was markedly higher in Temple
of Baal compared with the other caves, the highest
concentration of C0 2 occurred in the Chifley (5,232
ppm). With the exception of the Chifley and Temple
of Baal (maximum C0 2 = 3,662 ppm), the maximum
concentration of C0 2 that occurred at any of the other
sites within the caves was < 3,000 ppm (Fig. 2c).

Trends in cave atmosphere through time

Seasonal variation in the average daily air temp-
erature was most pronounced in the Chifley and

Site

River caves, with the difference in the average air
temperature between summer and winter of 0.8-1 ,2°C
in the Chifley and 0.6-0.7°C in the River Cave (Fig.
3). In the River Cave, the minimum air temperature
typically did not occur until the end of winter (mid-
late August), whereas in the Chifley, the minimum air
temperature occurred much earlier in winter. Seasonal
variation in the Imperial was inconsistent, with the
difference between summer and winter average air
temperature ranging from ~0.4°C in 2009 to ~0.1°C
2012 (Fig. 3). Unfortunately the temperature-relative
humidity unit in the Temple of Baal and data logger
in the Orient required repair on several occasions.

Proc. Linn. Soc. N.S.W., 136, 2014

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

(0, :i ) ajnjejsduuai

Proc. Linn. Soc. N.S.W., 136, 2014

Figure 3. Average daily air temperature at each of the monitoring sites within the caves.

A. BAKER

Consequently there are substantial gaps in the data
for both these caves, although the data suggest these
sites are characterised by generally highly stable air
temperatures.

With the exception of the Temple of Baal, the
concentration of CO, in the caves varied between
seasons. CO, peaked during the summer months
(December-January), decreased during autumn,
was at a minimum during winter (June-August)
and increased again in spring and summer (Fig. 4).
During summer, the maximum average concentration
of CO, in the caves was approximately 4-10 times
that of the external site, whereas in winter, the CO, in
the caves (excluding the Temple of Baal) was often
only slightly elevated above the external atmosphere,
with a maximum concentration 1.5-2 times that of
the external atmosphere. CO, in the Temple of Baal
showed no relationship with season, with major
increases typically occurring during periods of peak
visitation, as discussed in the following section.

Influence of visitors on cave atmosphere

Abrupt increases or “spikes” in air temperature
occurred in several caves and were most pronounced
in the Imperial, Lucas and Orient (Fig. 5a). Of these
caves, the Imperial experienced the largest spikes in
air temperature, with increases of up to 0.59°C in < 12
minutes during Easter 20 12 (Fig. 5a). Increases of 0.5-
0.7°C frequently occurred in the Imperial throughout
the four years of monitoring (as noted in the Bimonthly
reports prepared for JCRT), and this cave consistently
recorded the largest spikes in air temperature. The
largest single spike in air temperature, an increase of
0.9°C, occurred on the 10 th March 2012. Spikes of
between 0.2-0. 5°C were typical in the Lucas Cave,
as was evident during Easter 2012. Although spikes
in the Orient were often smaller than those in the
Imperial and Lucas (typically ~0.2°C), they generally
occurred on a more frequent and regular basis (Fig.
5a). It is important to note that tours in the Imperial
and Lucas frequently did not longer enter the sections
of the caves containing monitoring sites, whereas
every tour that entered the Orient entered the chamber
where monitoring was conducted.

With the exception of the Chifley Cave, the
RH in the other caves remained virtually constant
throughout the week, irrespective of commercial tours
(Fig 5b). In the Chifley Cave, RH, like air temperature
was highly variable and typically increased when air
temperature within the cave decreased, and decreased
when air temperature increased.

The concentration of CO, in cave atmosphere
generally increased midmorning each day, coinciding
with the commencement of cave tours. Increased

visitation over the Easter period had a pronounced
influence on the concentration of CO, (Fig. 5c). On
Thursday 5 th April, most of the caves recorded only
small increases in CO,, in keeping with relatively low
rates of visitation. Conversely over Easter, significantly
increases in visitation led to higher concentrations of
CO, for periods of time. For example in the Lucas
Cave, an increase in visitation of between 500-1,000
people/day during April 6-8 resulted in substantially
larger increases in CO, than April 5 (Fig. 5c), when
only 140 people visited the cave.

Different caves experienced different trends in
the accumulation of CO,. As with air temperature,
abrupt spikes in CO, occurred in the Imperial, Lucas
and the Orient (Fig. 5c). These abrupt increases were
usually relatively short-lived, however when tours
were frequent, the concentration of CO, did not
decrease to the pre-tour level before the next tour.
This frequently resulted in elevated levels of CO,
until there was a substantial gap between tours or
after the last tour for the day. In the River Cave, CO,
accumulation was more gradual, increasing mid-late
morning and decreasing each evening (Fig. 5c).

Interestingly, the Temple of Baal exhibited very
different trends in CO, compared to the other caves.
During periods of increased visitation such as Easter
2012, the level of CO, in the Temple of Baal gradually
increased with visitation during the day, plateaued or
marginally decreased during the evening and until the
commencement of tours the following day (Fig. 5c).
Conversely on days when there were fewer visitors,
such as April 5, 2012, there was a slight decrease in
the average concentration of CO,. Indeed throughout
2009-2012, periods of high visitation consistently
resulted in peak levels of CO, (Fig. 6).

To examine the relationship between visitation,
air temperature and CO, concentrations, these
parameters were graphed side by side for the Orient
Cave (Fig. 7). Simultaneous spikes in air temperature
and CO, corresponded with each and every one of the
tours through the cave. As could be expected, the size
of each tour influenced the magnitude of the spikes
in air temperature and CO,. For example on April
5, four similar sized tours during the day resulted in
four comparable spikes in air temperature and CO„
while a smaller 8 pm “extended Orient” tour resulted
in much smaller spikes in temperature and CO,.
During the Easter long weekend (April 6-9 2012), the
frequency of tours was such that after a tour, the air
temperature and CO, to did not decrease to the pre-
tour level before the next tour, resulting in a period
where temperature and CO, were elevated (Fig. 7).

Proc. Linn. Soc. N.S.W., 136, 2014

ooos

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

{uudd) apjxojp uoqjeo

-d

«d

-d

Proc. Linn. Soc. N.S.W., 136, 2014

A. BAKER

16.0

£ 13.5

13.0

-Chifley
Diamond
Imperial
•Lucas
Orient
River
• Temple of Baal

12.5

12.0
05- Apr- 12

06- Apr- 12

07- Apr- 12

08- Apr- 12

09- Apr- 12

10- Apr- 12

11 -Apr-1 2

100.0

99.5

99.0

| 98.5

£ 93.0

97.5

Chifley

Diamond

Imperial

Lucas

Orient

River

97.0 -I 1 1 i i 1 1

05- Apr- 12 06-Apr-12 07-Apr-12 08-Apr-12 09-Apr-12 10-Apr-12 11 -Apr-1 2

4000

3500

3000

2500

2000

1500

1000

500

0 -I 1 1 1 1 1 1

05- Apr- 12 06- Apr- 12 07-Apr-12 08- Apr- 12 09- Apr- 12 10- Apr- 12 11 -Apr-1 2

Date

^—Chifley

Diamond

Imperial

Lucas

Orient

River

— Temple of Baal
External

Figure 5. Daily variation in a) air temperature, b) relative humidity (RH) and c) C0 2 concentration at
each of the monitoring sites from April 5 to 11 2012. Each parameter was recorded at 6 minutes inter-
vals.

Proc. Linn. Soc. N.S.W., 136, 2014

4000 n r 300

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

Number of visitors

(uidd) epjxojp uoqjeo

Proc. Linn. Soc. N.S.W., 136, 2014

Figure 6. Relationship between the concentration of carbon dioxide in the Temple of Baal and number of visitors to the cave each day.
Note: no visitor data were available for March 23-29 2010.

17.0 i T 2000

A. BAKER

(Oo) ainiejadiuej.

sjojsjA jo jaquinN

* 3 S

(Z3

V-

a a>
o u

£ -53
S

» S3

£ 9 j 6fj

3 ^

W> ’S

EiS

S3 «®

« 3

e 4

s S3 *-
© S3
CS W 0)

|§g

a a o

Proc. Linn. Soc. N.S.W., 136, 2014

Date

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

DISCUSSION

Trends in cave atmosphere

Of all the caves, the Chill ey had the most variable
atmosphere, and experienced considerable day to
day variation and pronounced seasonal trends in air
temperature and CO,. These trends are consistent
with strong ventilation in the cave (Whittlestone
et al. 2003) and current research by Waring and
Hankin (2013), which has shown the occurrence of
bi-directional airflow and daily cycles of ventilation
through the upper (i.e. the Elder / Plug Hole) and lower
entrances to the cave. Conversely, the most stable air
temperature occurred in the Temple of Baal, a cave
known to have limited ventilation (Whittlestone et al.

2003) . As with air temperature, strong ventilation in
the Chifley Cave accounts for greater variability of
RH in the Chifley compared to the other caves.

The average (2009-2012) air temperature differed
by up to 2.8°C between the sites within the caves.
These differences were predominately attributed to
the location of individual caves within the larger cave
system. For example the warmest caves were the
Temple of Baal, the Orient and Mafeking branch of
the Lucas Cave, all of which are situated in the upper
levels of the southern show caves, while the River
Cave at a lower level was substantially cooler. In the
northern show caves, the average air temperature in
the Diamond was slightly cooler than the Chifley and
Imperial. On the basis of elevation, this result would
not be expected, since the Diamond is higher than
the Imperial. However connections to other passages
(e.g. the Jubilee Cave) create complex airflows
through the cave (J. James pers. com.), which may
explain this discrepancy. More broadly, since the air
temperature of a cave is frequently influenced by the
surface temperature of the ground, it is plausible that
variations in the microclimate of the surface karst
(e.g. due to differences in aspect, exposed bedrock
and vegetation cover) contributed to the difference in
air temperature between the caves (Dominguez- Villar
et al. 2013)

With the exception of the Temple of Baal, all of
the caves exhibited seasonal variation in CO„ with
a summer maximum and a winter minimum. These
trends are consistent with Australian (e.g. Smith
1998; Eberhard et al. 2003) and international (Spotl
et al. 2005; Fernandez-Cortes et al. 2006; Linan et al.
2008) literature which report that CO, concentrations
typically peak in summer and decrease to a minimum
in winter. Although changes in barometric pressure
can cause air to flow into or out of a cave (Cigna

2004) , external air temperature is a likely driver of
seasonal variation in CO, at Jenolan. During winter,

the cave atmosphere is generally substantially warmer
than the external air, which increases air circulation as
warm, moist air rises out of the cave and is replaced
by comparatively cold, dry external air. Conversely,
when external temperatures are warmer, a temperature
inversion does not form and the comparatively cool,
moist air remains within in the cave (Fernandez-
Cortes et al. 2006). This presumption is in accordance
with contemporary research by Waring and Hankin
(2013), who found external air temperature is the key
driver of air circulation in the Chifley Cave.

Furthermore, higher concentrations of CO,
during summer are also likely, due to increased
production of CO, by natural processes (Baker and
Genty 1998). These include the diffusion of C0 2 from
within epikarst air (rich in CO, from root respiration
and the breakdown of organic matter), degassing from
cave waters, and biological productivity in the cave
(Fairchild and Baker 2012). Indeed contemporary
research by Waring and Hankin (2013) has shown
that on hot days, the airflow in the Chifley Cave
causes soil-air that is rich in CO, to seep into the cave,
particularly at Katies Bower.

In the Temple of Baal, the gradual accumulation
of CO, during periods of high visitation, gradual
decrease in CO, during periods of lower visitation
and lack of seasonal variation was indicative of low
ventilation during 2009-20 1 2 . This notion is consistent
with the findings of Whittlestone et al. (2003), who
reported low ventilation rates, such that the difference
between the summer and winter concentrations of
radon in the Temple of Baal was only “marginally
discernable” (Whittlestone et al. 2003). The increases
and decreases in CO, during periods of high and low
visitation confirm that visitation is an important factor
in the determining CO, levels within this cave and
is consistent with the hypothesis of Michie (1997),
that visitors to the Temple of Baal caused the high
concentrations of CO,.

Implications for management

The Chifley was the only cave where the
concentration of CO, exceeded 5,000 ppm, the limit
specified in Carbon Dioxide Exposure Standards (time
weighted average), which allow a person to work an 8
hour day for 40 hours per week in 5,000 ppm of CO,
(Safe Work Australia 2013). The same standards also
specify a short term exposure limit of 30,000 ppm (3
%) CO, for a duration of 15 minutes. Although the
maximum concentration of CO, exceeded 5,000 ppm
in the Chifley, this was never for an extended period
of time, as evident by the average daily concentration,
which only once exceeded 4,500 ppm and never 5,000
ppm (Fig. 4). Consequently, it is extremely unlikely

Proc. Linn. Soc. N.S.W., 136, 2014

A. BAKER

that the concentrations of CO, reported at any of the
monitoring sites would have impacted the health or
safety of visitors, cave guides, or maintenance staff

Commercial cave tours frequently increased the
air temperature and concentration of CO, within the
cave atmosphere. In most of the caves these increases
were relatively short lived and rapidly returned to the
pre-tour level (i.e. fast relaxation time) immediately
after the tour when visitation was moderate or at
the end of the day during periods of high visitation.
Consequently, in these caves, the current rates of
visitation and tour schedules did not have any apparent
lasting impact on the cave atmosphere. However the
same was not true for the Temple of Baal, where the
number of visitors influenced the concentration of
CO, and resulted in an accumulation of CO, during
periods of high visitation.

CO, is a crucial factor in many of the processes
that form caves and their speleothems. An increase in
the concentration of CO, in the cave atmosphere may
initially impact active speleothems by decreasing
the rate of calcite deposition and ultimately the
dissolution of speleothems (James 2004). Kermode
(1979) proposed that concentrations of CO, above
2,400 ppm result in aggressive water that can
dissolve speleothems, and as a result, is the maximum
permissible level in Glowworm Cave, New Zealand
(de Freitas and Banbury 1999; de Freitas 2010),
although the reliability of this threshold as a universal
guideline has been questioned (e.g. Michie 1997; de
Freitas and Banbury 1999). Recent research suggests
that there is no universal threshold, but rather that
the equilibrium of CO, between the air and water
ultimately determines if calcite is deposited or
dissolved (Baker and Genty 1998; Cigna 2002; James
2004). Moreover, research examining the influence of
CO, on calcite deposition within the tourist caves at
Jenolan found corrosion thresholds ranged from 2,690
ppm in the River Cave to 28,000 ppm in the Ribbon
Cave and did not exceed the maximum CO, measured
in the caves (Failes 1997). It is important to note that
these thresholds do not apply to inactive speleothems
and bedrock, and corrosive condensates that form
as a result of increased concentrations of CO, from
visitors can be highly damaging (James 2004, 2013),
although fortunately many of the speleothems in the
Temple of Baal appear to be active.

It is beyond the scope of the JEMP to measure
and quantify corrosion thresholds of C0 2 within the
caves and further research on the impacts of CO,
on the caves is required, especially in the Temple
of Baal. Specifically, measurement of the partial
pressure of CO, (P co ,) of speleothem drip water and
concentration of dissolved calcium could be used to

accurately determine the sensitivity of speleothems
to changes in the concentration of CO, in the cave
atmosphere (Fairchild and Baker 2012). Nevertheless,
in the Temple of Baal, visitation increased the
concentration of CO, for lengthy periods of time,
such that no seasonal variability was evident, and may
in turn influence process such as the rates of calcite
deposition. Furthermore, although the concentration
of CO, in the Temple of Baal was below the limit
specified by Safe Work Australia (20 1 3), which allows
a person to work an 8 hour day (40 hour week) in
5,000 ppm of CO,, it has been suggested that visitors
may experience discomfort from CO, concentrations
< 2,500 ppm (Osborne 1981).

The most common methods of addressing issues
of air quality in tourist caves are to limit visitor
numbers and artificial ventilation (James 2004).
Obviously if visitation causes an accumulation of
CO, in a cave, a reduction in the number of people
who visit that cave will lessen this accumulation.
However, this is in direct conflict with commercial
interests and is not a considered a realistic proposition.
At the same time, given that visitors are elevating the
concentration of CO, in the Temple of Baal, caution
should be exercised when evaluating visitation rates
and the possibility of increasing the number of people
who visit the cave.

When considering the influence of visitation
on the atmosphere and microclimate of Pozalagua
Cave (Spain), Lario and Soler (2010) recommended
closure of the cave one day per week during periods of
“normal” visitation and two days per week after high
visitation in order to minimise the cumulative effects
of visitation. Given the low rates of ventilation and
time taken for the concentration of CO, to decrease,
it is unclear how successful a similar scenario would
be in the Temple of Baal. Under the existing trends
in visitation, the concentration of CO, gradually
decreases with lower visitation following periods of
peak visitation and accordingly, the environmental
benefits of completely closing the cave would need
to be weighed up against the economic benefits
associated with visitation.

An alternative method of dealing with CO, is to
ventilate the cave to prevent a build up of excessive
CO,. For example, careful manipulation of the
airflow regimes is used to limit the accumulation of
CO, in Glowworm Cave, New Zealand (de Freitas
and Banbury 1999; de Freitas 2010; Gilles and de
Freitas 2013). Similarly, Michie (1997) demonstrated
that opening the airtight doors in the Binoomea Cut,
an artificial tunnel that provides access to the Temple
of Baal, can rapidly decrease the concentration of
CO, in the cave atmosphere. However increased

Proc. Linn. Soc. N.S.W., 136, 2014

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

ventilation is likely to cause significant side effects
that must be carefully considered. The maintenance
of natural conditions is crucial to the conservation
of a cave (Watson et al. 1997) and changes to the
natural airflow frequently alter the microclimate of
a cave (Gillieson 1996). Artificial ventilation often
causes increased fluctuations in temperature and
relative humidity (Russell and Maclean 2008) and is
a major cause of desiccation of caves (Gillieson 1996;
de Freitas 1997). Additionally, increased airflow
circulates dust particles to a greater depth within
the cave (Michie 2004) and is likely to discolour
speleothems, thereby reducing their aesthetic value
(James 2013). Consequently, any potential change to
the air flow in a cave is potentially highly damaging
and requires careful consideration (Michie 2004;
Faimon et al. 2012).

The creation of artificial entrances modifies
natural airflow, thereby altering the natural
microclimate of a cave (Cigna 1993; Gillieson

1996) . For this reason the International Show Cave
Association (ISCA) states that “any new access into
a cave must be fitted with an efficient system, such as
double set of doors, to avoid creating changes in the
air circulation” (ISCA 2010). The artificial entrance
to the Temple of Baal, the Binoomea Cut, contains
two air lock doors, which were installed after it was
observed that the cave was drying out (J. James
pers. com.). Prolonged opening of the airlock doors
would undoubtedly increase airflow and decrease in
CO, but is also likely result in the desiccation of the
cave, especially in winter, when the artificial entrance
may act as a “chimney” (see de Freitas and Banbury
1999 and Russell and Maclean 2008) as wann moist
air is drawn out of the Binoomea Cut, while cold
air is drawn in from the River Cave. This scenario
would be highly undesirable, as the potential befits
associated with decreased levels of CO, would almost
certainly be outweighed by unnatural variation in the
microclimate and desiccation of speleothems.

Previous studies have noted the conflict between
maintaining a stable microclimate (in particular
temperature and RH) versus the need for ventilation
to manage the accumulation of CO, (e.g. de Freitas
1997; Michie 1999; Linan et al. 2008). One solution
to this conflict may be a compromise whereby limited
ventilation is permitted through the Binoomea Cut.
This could be achieved by temporary opening of
the airtight doors (e.g. two hours as per Michie

1997) , although this would be expected to increase
the variation of air temperature and RH in the cave.
Additionally, the manual opening and closing of the
doors, may be problematic as it could be expected
that the doors would accidently be left open from time

to time, to the detriment of the cave. An alternative
approach could be the installation of a window in
each of the airlock doors that could be opened during
peak visitation to allow limited airflow into the cave.
A more limited airflow may allow the temperature
and RH of the external air to partially equalise with
the cave atmosphere before reaching the cave, whilst
minimising the potentially harmful accumulation of

co,.

As previously discussed, the circulation of air in
caves is influenced by a number of factors including
season and weather conditions. These factors may
have significant implications for ventilation, since
the influence of ventilation on the cave microclimate
as well as its effectiveness in removing CO, can
be highly variable depending on season and local
weather conditions, and require different ventilation
regimes (de Freitas 1997). With this in mind, any
study of the effectiveness and impacts of ventilation
must include temporal variation in airflow associated
with season and varying weather conditions. Finally,
in considering the possibility of increasing the
ventilation in caves such as the Temple of Baal, it
must stressed that any change to airflow within a
cave requires careful consideration, and must be
guided by ongoing monitoring, if long tenn impacts
are to be minimised. Such considerations highlight
the importance and value of long term, baseline data
collected in environmental monitoring programs such
as the Jenolan Environmental Monitoring Program.

Monitoring of the cave atmosphere at Jenolan
provides valuable baseline data for the air temperature,
relative humidity and concentration of CO, in the tourist
caves. Regular measurement of these parameters has
ascertained the caves are typically characterised by
high levels of relative humidity, moderately stable
air temperature with seasonal variation of < 2°C, and
highly seasonal variation in concentration of CO,.
Commercial cave tours frequently increased the air
temperature and concentration of C0 2 , although both
parameters rapidly returned to the pre-tour level after
the tour had passed. An exception occurred in the
Temple of Baal, where peak visitation elevated the
concentration of CO, for extended periods of time,
such that seasonal variation was not apparent.

ACKNOWLEDGMENTS

Many people in NSW Office of Environment and Heritage
and Jenolan Caves Reserve Trust have played a crucial role
in the JEMP. In particular, thanks to Stephen Meehan (OEH)
for his efforts in establishing and managing the JEMP and
Russell Cummins (formerly OEH) for his preliminary work

Proc. Linn. Soc. N.S.W., 136, 2014

A. BAKER

during the establishment of the Program. Thanks also to Dan
Cove and Grant Cummins (JCRT) for the logistical support
at Jenolan. Dr. Julia James kindly commented on a draft
of the manuscript. Prof. Andy Baker and an anonymous
reviewer provided valuable comments on the manuscript.

REFERENCES

Baker, A. and Genty, D. (1998). Environmental pressures
on conserving cave speleothems: effects of changing
surface land use and increased cave tourism. Journal
of Environmental Management 53 , 165-175.

Cigna, A.A. (1993). Environmental management of tourist
caves. Environmental Geology 21 , 173-180.

Cigna, A.A. (2002). Modern trend in cave monitoring.
Acta carsologica 31 , 35-54.

Cigna, A.A. (2004). Climate of Caves. In ‘Encyclopaedia
of caves and karst science’ (ed. J. Gunn) pp. 228-230.
(Fitzroy Dearborn: New York).

Department of Environment and Climate Change (2009).
‘Greater Blue Mountains World Heritage Area:
strategic plan’. (Department of Environment and
Climate Change (NSW)).

de Freitas, C.R. (1997). Cave monitoring and

management: The Glowworm Cave, New Zealand.

In: ‘Proceedings of the 12th ACKMA Conference,
Waitomo, NZ, 1997.’ (ed. D.W. Smith) pp. 55-
66. (Australasian Cave and Karst Management
Association: Carlton South, Victoria, Australia).

de Freitas, C.R. (2010). The role and importance of cave
microclimate in the sustainable use and management
of show caves. Acta carsologica 39 , 477-489.

de Freitas, C.R. and Banbury, K. (1999) Build up and
diffusion of carbon dioxide in cave air in relation
to visitor numbers at the Glowworm Cave, New
Zealand. In: ‘Proceedings of the Thirteenth
Australasian Conference on Cave and Karst
Management, Mount Gambier, South Australia,
1999.’ (eds. K. Henderson and P. Bell) pp. 84-
89. (Australasian Cave and Karst Management
Association: Carlton South, Victoria, Australia).

Eberhard, S., Michie, N., McBeath, R. and Bell, P.

(2003). Carbon dioxide and climate in Jewel Cave.

In ‘Proceedings of the Fifteenth Australasian
Conference on Cave and Karst Management,
Chillagoe Caves and Undara Lava Tubes, North
Queensland, 2003. ’(ed. K. Henderson) pp. 73.
(Australasian Cave and Karst Management
Association Incorporated: Carlton South, Victoria,
Australia).

Failes, T. (1997). Carbon dioxide and its influence on
calcite deposition and solution. Honours Thesis,
University of Sydney, Sydney.

Faimon, J., Troppova, D., Baldik, V. and Novotny,

R. (2012). Air circulation and its impact on
microclimatic variables in the Cisarska Cave
(Moravian Karst, Czech Republic). International
Journal of Climatology 32 , 599-623.

Fairchild, I.J. and Baker, A. (2012) ‘Speleothem science:
From processes to past environments’ Wiley -
Blackwell, Oxford, UK.

Femandez-Cortes, A., Calaforra, J.M. and Sanchez-
Martos, F., (2006). Spatiotemporal analysis of
air conditions as a tool for the environmental
management of a show cave (Cueva del Agua,

Spain). Atmospheric Environment 40 , 7378-7394.

Gilles, M.J. and de Freitas, C.R. (2013). Environmental
Management of the Waitomo Glowworm Cave:
Effects of visitors and ventilation on carbon dioxide
concentrations. In ‘Proceedings of the Fifteenth
Australasian Conference on Cave and Karst
Management, Waitomo Caves, New Zealand, 2013’
(eds R. Webb and S Webb) pp 173-191. (Australasian
Cave and Karst Management Association
Incorporated: Mount Compass, South Australia,
Australia).

Gillieson, D. (1996) ‘Caves: processes, development and
management’. Blackwell, Oxford, England.

International Show Cave Association (2010) ISC A
management guidelines - draft. http://www.i-s-c-
a. com/documents

James, J.M. (2004). Tourist caves: air quality. In

‘Encyclopaedia of caves and karst science’ (ed. J.
Gunn) pp. 730-731. (Fitzroy Dearborn: New York).

James, J.M. (2013). Atmosheric processes in caves.

In ‘Treatise on Geomorphology, Vol 6, Karst
Geomorophology’ (Editor in Chief J.F.Shroder,
Volume Editor A. Frumkin) pp 304-318. (Acedemic
Press, San Diego).

Jenolan Caves Reserve Trust (2012). Jenolan Caves
Reserve Trust: annual report 2011-2012. http://
www.ienolancaves.org.au/imagesDB/wvsiwvg/
AnnualReport2011-12.pdf

Kermode, L.O. (1979). Cave corrosion by tourists. In
‘Cave Management in Australia III: Proceedings
of the third Australian conference on cave tourism
and Management, Mt Gambier, South Australia,
1979.’ pp. 97-104. (Australasian Cave and Karst
Management Association, Carlton South, Victoria,
Australia).

Lario, J. and Soler, V. (2010). Microclimate monitoring
of Pozalagua Cave (Vizcaya, Spain): application to
management and protection of show caves. Journal
of Cave and Karst Studies 72 , 1 69-1 80.

Linan, C., Vadillo, I. and Carrasco, F. (2008). Carbon
dioxide concentration in air within the Nerja Cave
(Malaga, Andalusia, Spain). International Journal of
Speleology 37, 99-106.

Michie, N. (1997). An investigation of the climate, carbon
dioxide and dust in Jenolan Caves, NSW. PhD thesis,
Macquarie University, Sydney.

Michie, N. (1999). Management of carbon dioxide in
show caves. In ‘Cave Management in Australasia
13: Proceedings of the 13 th ACKMA Conference,

Mt Gambier, South Australia, 1999.’ (eds K.
Henderson and P. Bell). (Australasian Cave and Karst
Management Association, Carlton South, Victoria,
Australia).

Proc. Linn. Soc. N.S.W., 136, 2014

ENVIRONMENTAL MONITORING AT JENOLAN CAVES

Michie, N. (2004). Tourist caves: air debris. In

‘Encyclopaedia of caves and karst science’ (ed. J.
Gunn) pp. 731-732. (Fitzroy Dearborn: New York).

Osborne, R.A.L. (1981). Towards an air quality standard
for tourist caves: studies of carbon dioxide enriched
atmosphere in Gaden-Coral, Wellington Caves, NSW.
Helictite 19 , 48-56.

Pulido-Bosch, A., Martin-Rosales, W., Lopez-Chicano, M.,
Rodriguez-Navarro C.M., and Vallejos, A. (1997).
Human impact in a tourist karstic cave (Aracena,
Spain). Environmental Geology 31 , 142-149.

Russell, M.J. and MacLean, V.L. (2008). Management
issues in a Tasmanian tourist cave: Potential
microclimatic impacts of cave modifications. Journal
of Environmental Management 87 , 474-483.

Safe Work Australia (2013) Hazardous Substances
Information System

http://hsis.safeworkaustralia.gov.au/ExposureStandards

Smith, G.K. (1998). Foul air at Bungonia. In ‘Under
Bungonia’ (eds J., Bauer and P, Bauer) pp. 85-92.

(JB Books: Oak Flats, NSW, Australia).

Spotl, C., Fairchild, I J. and Tooth, A.F. (2005). Cave
air control on dripwater geochemistry, Obir Caves
(Austria): Implications for speleothem deposition
in dynamically ventilated caves. Geochimica et
Cosmochimica Acta 69 , 2451-2468.

Waring, C.F. and Hankin, S.I. (2013). How weather and
climate drives Chifley Cave speleothem growth. In
‘The Science of Jenolan Caves: What do we know?’
Abstracts of presented papers and posters (Jenolan
Caves, May 23-24, 2013).

Watson, J., Hamilton- Smith, E. Gillieson, D. and Kieman
K. (eds) (1997). ‘Guidelines for cave and karst
protection’ . (International Union for Conservation of
Nature and Natural Resources: Cambridge, UK).

Whittlestone, S., James, J. and Bames, C. (2003). The
relationship between local climate and radon
concentrations in the Temple of Baal, Jenolan Caves,
Australia. Helictite 38 , 39-44.

Proc. Linn. Soc. N.S.W., 136, 2014

Invertebrate Cave Fauna of Jenolan

Stefan M. Eberhard 1 , Graeme B. Smith 2 , Michael M. Gibian 3 , Helen M. Smith 4 and

Michael R. Gray 4

1 Subterranean Ecology Pty Ltd, Perth; 2 Australian Museum Entomology Dept., 6 College St, Sydney
2010 and author for correspondence (le_gbsmith@optusnet.com.au); 3 SC Johnson and Son Inc., Sydney;

4 Australian Museum Arachnology Dept.

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/joumals/index.php/LIN

Eberhard, S.M., Smith, G.B., Gibian, M.M., Smith, H.M. and Gray, M R. (2014). Invertebrate cave fauna
of Jenolan. Proceedings of the Linnean Society of New South Wales 136 : 35-67.

The invertebrate fauna known from within the caves at Jenolan is inventoried and summarised. At least
136 individual taxa have been identified although less than one-half (43%) are assigned to described
species, the rest are either undescribed (8%) or have only been identified to genus level (31%) or higher taxa
(18%). The collected fauna is dominated by arachnids (47%) and collembolans (24%) followed by insects
(15%) and crustaceans (6%) with three or fewer taxa identified in each of the remaining groups comprising
molluscs, diplopods, chilopods, annelids, platyhelminths and nematodes. In terms of ecological dependence
on caves, 53% of collected taxa comprised typically epigean species with the remainder considered to be
habitual cave-dwellers. Eight species (revised from 14 previously) are considered to be obligate hypogean
species (terrestrial troglobites or aquatic stygobites) comprising three species of springtail, two spiders, a
pseudoscorpion and two aquatic crustaceans. The diversity of troglobite species is fairly typical for karst
areas in the eastern highlands of NSW but higher unrecorded diversity of stygobite species is predicted.

While the invertebrate cave fauna of Jenolan has received more attention from biologists than any other
karst area in NSW, substantial knowledge gaps remain. Research and conservation priorities are: (1)
identify existing collections and describe new species, focussing on troglomorphic taxa which are likely
to be locally endemic and of conservation significance; (2) targeted held surveys for rare troglomorphic
taxa which are under-represented in existing collections; (3) sample for aquatic micro-cmstacea and other
stygofauna in vadose zone, phreatic zone and interstitial habitats; (4) sample for troglobites in meso-cavern
and other cryptic terrestrial habitats.

Manuscript received 23 October 2013, accepted for publication 11 December 2013.

KEYWORDS: cave fauna, Jenolan, stygobite, troglobite

INTRODUCTION

The purpose of this paper is fourfold: (1) to
provide an historical inventory of the invertebrate
cave fauna recorded from the Jenolan karst, which
to date, has largely existed in unpublished reports;
(2) to summarise the current state of taxonomic and
collection knowledge; (3) to identify knowledge gaps
and priorities for further research and conservation;
(4) to briefly re-assess the significance of the Jenolan
cave fauna in a regional and national context.

The Jenolan Caves have attracted the attention of
Europeanscientistssincehrst being visited in the 1 830s,
however little attention was paid to the invertebrate
fauna, either above or below ground, until guide
Joseph C. Wiburd initiated collections from the 1 880s

until around 1903. Many ofWiburd’s specimens are in
the Australian Museum collections. Most specimens
appear to be surface collections although two species
of cave-dwelling spider ( Cycloctenus abyssinus and
Laetesia weburdi ) described by Urquhart (1890), are
historically important, being the first cave dwelling
invertebrates described from New South Wales.

After Wiburd and Urquharf s pioneering efforts,
further documentation of Jenolarfs invertebrate
cave fauna lapsed until the 1960s when collections
were reinitiated by John Polesson, Barbara Dew,
Elery Hamilton- Smith, Ted Lane and Aola Richards.
Their efforts identified ten named species of spider,
pseudoscorpion, harvestman, springtail and beetle,
plus several other unidentified species of millipede,
cricket and moth (Hamilton- Smith 1967).

The next era of systematic survey occurred

INVERTEBRATE CAVE FAUNA OF JENOLAN

between 1986 and 1988, when Michael Gibian, Louise
Wheeler and Graeme Smith, with further involvement
from Mike Gray, Glenn Hunt, Penelope Greenslade,
Mia Thurgate and Ernst Holland, sampled the fauna
by hand as well as netting streams and taking samples
of leaf litter and guano for Tullgren funnel extractions.
These efforts increased the number of recorded taxa
(most undescribed) from 26 to 67 including Jenolan’s
first troglobitic spiders and aquatic cave fauna (Gibian
et al. 1988).

Systematic collection efforts were continued by
Eberhard ( 1 993) with emphasis on aquatic macrofauna
and interstitial habitats using baits, nets and pumping
methods. These collections and other previous
accessible records were part of a wider survey of New
South Wales cave fauna which established Jenolan
as one of the better sampled karsts in the State and
possessing a comparatively rich invertebrate cave
fauna (Eberhard and Spate 1995). Since this last
survey and inventory at Jenolan, which remains
unpublished in the scientific literature, further field
collection efforts have been very limited.

As is typical of invertebrate surveys, and
subterranean fauna especially, the taxonomic
(Linnaean) shortfall means that much of the Jenolan
material remains incompletely identified, awaiting
specialist attention. Some progress has however been
made with descriptions of four mite species (Halliday
2001), one spider (Forster et al. 1987), one amphipod
(Bradbury and Williams 1997), redescription of
the Jenolan harvestman (Hunt 1992), and further
identification of springtails (Greenslade 2011);
descriptions of an additional four mite species are in
preparation (Halliday in litt. 2013).

The survey and inventory by Eberhard and Spate
( 1 995) informed the stance taken in a subsequent paper
by Thurgate et al. (2001a) who applied the metaphor
‘from rags to riches’ to highlight subterranean
biodiversity in New South Wales and ‘dispel former
erroneous perceptions of a depauperate fauna’. Since
this paper was published, a great amount of field
survey and taxonomic research has been undertaken
in other states, mostly in Western Australia and South
Australia (Eberhard et al. 2009; Guzik et al. 2011),
the results of which reinforce the need and timeliness
for formal documentation and reappraisal of Jenolan’s
cave fauna as presented herein.

DEFINITIONS

Biospeleologists classify subterranean species
according to their degree of ecological association
and dependence upon subterranean environments.

Frequently this association is presumed or inferred,
especially in the case of obligate subterranean
forms, on the basis of morphological modifications,
typically a reduction or loss of pigmentation and
eyes, elongation of appendages and compensatory
enhancement of non-optic sensory structures.

Accidentals: Typically surface dwelling

species whose occurrence underground is
incidental, having ‘accidentally’ wandered
or fallen in, or been carried underground
by sinking water (e.g. flood), gravity or air
currents

Epigean: Surface dwelling

Hypogean: Subterranean

Guanophile/Guanobite: Species that are

associated with the guano of cave roosting
bats or birds. Species associations with
guano may be facultative (guanophile) or
obligate (guanobite).

Meso-cavem: Subsurface cavity generally too
small for a human to enter. Underground
voids in the size range 0.1-20 cm,
especially in karst and volcanic substrates,
cf. macro-cavern which are voids > 20 cm,
especially caves large enough for human
entry.

Stygophile/Stygobite: Terms equivalent to troglo-
phile and troglobite for aquatic cave fauna

Trogloxene: Species that habitually occupy caves
for a part of their life cycle but frequently
return to the surface for food. e.g. bats and
cave crickets.

Troglophile: Species that can complete their

whole life cycle in hypogean environments
but populations of the same species
also occur in epigean environments.

They usually do not possess typical
morphological modifications, but in some
cases the cave-dwelling populations
may show some degree of modification
(e.g. lighter pigmentation or reduced eye
size) compared to their surface-dwelling
conspecifics.

Troglobite: Species that are obligate cave
dwellers and entirely restricted to the
subterranean environment and showing
typical troglomorphic traits (see next).

Troglomorphy: Any morphological, physio-
logical, or behavioural feature that
characterizes subterranean fauna. Common
morphological traits include: reduction
of eyes, pigment, wings; elongation of
appendages; specialization of non-optic
sensory structures.

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

OVERVIEW

An overview of the systematic composition and
current state of taxonomic knowledge appears in Fig.
1 and Table 1 and a more comprehensive list of the
faunal records and the location of specimens is in the
appendix. At least 136 individual taxa have so far
been collected within the caves at Jenolan. In terms
of recorded diversity, the collected invertebrate fauna
is dominated by arachnids (47%) and collembolans
(24%) followed by insects (15%) and crustaceans
(6%) with three or fewer taxa identified in each of the
remaining groups comprising molluscs, diplopods,
chilopods, annelids, platyhelminths and nematodes
(Fig. 1).

Springtails (Collembola) were very abundant
and diverse with 33 recognised taxa including
three troglobites and seven undescribed species
(Table 1). Although a naturally diverse group, their
disproportionate representation in Jenolan cave
collections partly reflects the survey and identification
efforts applied to this group by Greenslade (2002)
and which contrasts with most of the insect groups
excepting the beetles (Coleoptera) which are

reasonably well known. The arachnid collections
are dominated by terrestrial mites (Acarina) and
spiders (Araneae) with 28 and 31 recognised taxa
respectively. This also partly reflects the survey and
identification efforts for these groups applied by
Halliday (2001) and Gray (1973) respectively. While
eight crustacean taxa have been recorded to date, this
is likely to under-represent the actual diversity because
this group is typically diverse in karst groundwater.
Moreover, Jenolan’s deep groundwater habitats have
been poorly sampled for aquatic micro-crustacea. In
terms of taxonomic resolution, less than one-half (59
species, 43%) of the 136 taxa are currently assigned
to described species, the rest are either undescribed
(11 species, 8%) or have only been identified to genus
level (42 taxa, 31%) or higher (24 taxa, 18%) (Table
1 ).

A systematic list of all invertebrate taxa recorded
from inside caves at Jenolan is given in the appendix.
In terms of ecological classification, many of the
taxa are considered to be ‘accidental’ or incidental
hypogean fauna (72 taxa), falling into caves or being
washed in by flood events. Forty-nine (49) taxa are
considered to be troglophiles (or stygophiles). Only

ANNELIDA, 1

DIPLOPODA, 1

ENTOGNATHA, 34

CHILOPODA, 1

CRUSTACEA, 8

INSECTA, 21

MOLLUSCA, 3

PLATYHELMINTHES, 2

NEMATODA, 1

ARACHNIDA, 64

Figure 1. Systematic composition of Jenolan invertebrate cave fauna collections showing the number of
taxa identified in major taxonomic groups.

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

Table 1. Overview of recorded diversity and taxonomic resolution in major selected groups of
Jenolan cave invertebrates.

Higher Group

No.

taxa

Described

sp.

Undescribed
n. sp.

Identified to
genus

Not identified
to genus

Troglobites /
stygobites

Entognatha:

Collembola

Entognatha:

Other

Insecta:

Coleoptera

Insecta: Others

Arachnida:

Araneae

Arachnida:

Acarina

Arachnida:

Others

Crustacea

Diplopoda

Chilopoda

Mollusca

Annelida

Nematoda

Platyhelmin-

thes

Totals

136

eight species are considered to be troglobites or
stygobites, comprising three species of springtail,
two spiders, a pseudoscorpion, and two crustaceans
(Table 1, Figs 2, 3 and 4).

DETAILED SYSTEMATIC ACCOUNT WITH
NOTES ON COLLECTIONS AND ECOLOGY

ENTOGNATHA

Subclass Collembola

Penelope Greenslade has tentatively identified
33 taxa from 1 1 families from material predominantly
collected by Gibian, Smith, Wheeler, and Eberhard
(Greenslade 2002). Collembola were mainly
collected by hand from the surface of pools, from
rock walls, stalagmites and other surfaces, but some
Tullgren funnel extractions were taken of guano

and flood debris, and some pitfall traps baited with
arthropod remains. Collembola were observed to be
very abundant on moist surfaces (e.g. stalagmites) in
the humid and dark sections of caves developed for
tourism (e.g. Orient Cave upper levels) (S. Eberhard
personal observation, 1993). It is hypothesised that
tourism activities have altered the ecology of these
otherwise normally dark and energy-poor deep zone
environments, via the introduction of artificial light
and nutrients with associated growth of fungi and
lampen-flora which provide a food source for grazing
invertebrates to colonise deep zone habitats that
would normally preclude them.

The most abundant species {Onychiurus sp.
fimetarius group, Ceratophysella spp. Mesophorura
sp. krausbaueri group and Folsomia Candida (Willem,
1902)) also occur in Europe and are almost certainly
introduced to Australia. The undescribed native
Adelphoderia sp. was the most frequently occurring

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

Figure 2 (left). Scanning electron
micrograph of Adelphoderia sp.,

< 1 mm (Penelope Greenslade)

Figure 3 (below). Examples of Jeno-
lan cave fauna, approximate length
(including legs) indicated (photog-
rapher). a. Cavernotettix cave crick-
et, 25 mm (Stefan Eberhard); b. Ba-
dumna socialis 16 mm (Mike Gray);
c. Stiphidion facetum (with dipteran
prey), 25 mm (Stefan Eberhard); d.
Web of S. facetum (Helen Smith);
e. Laetesia weburdi , 5 mm (Mike
Gray); f. Holonuncia cave harvest-
man, 20 mm (Stefan Eberhard).

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

Figure 4. Examples of Jenolan cave fauna, a. Trechimorphus diemenensis , 5 mm; b. Pseudoscorpion
Sathrochthonius tuena , 1.4 mm; c. Troglobitic pseudoscorpion Pseudotyrannochthonius jonesi , 3 mm; d.
Icona sp., 8 mm, a troglophile with pigment and eyes; e. Troglobitic Theridiidae sp. (previously as Icona
sp. 3), 3mm; f. Stygobitic amphipod Neocrypta simoni, 4 mm; g. Stygobitic crustacean, Psammaspididae
gen. et sp. nov. 5mm (a.- f. Mike Gray; g. Peter Serov).

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

species (Fig. 2). Almost half the number of taxa were
recorded only once or twice, mostly from extractions
of flood debris and are almost certainly ‘accidentals’
washed in by flood waters. The Jenolan fauna was
found to contain a greater number of genera with
exotic species compared with the Tasmanian cave
fauna (Greenslade 2002).

Greenslade considered that four of the Jenolan
species were likely troglobites and another 10 species
probable troglophiles. The troglobitic species of
most interest from conservation and phylogenetic
points of view {Kenyura sp.) is known, to date, only
from a single cave. With the exception of Coecobrya
communis (Chen and Christensen 1997) (an exotic
introduced species previously incorrectly identified
as Lepidosinella armata ), none of these species
has yet been described. Coecobrya communis was
later reported by Chen et al (2005) to also occur in
worm beds and is therefore considered in this work
as a troglophile rather than a troglobite. Within the
Jenolan Caves it has been collected from drains and
gutters and on stalagmite.

All troglobitic Collembola, except Adelphoderia
sp., were rare in the collections. Kenyura sp. was
collected from mud banks and the surface of muddy
pools; Oncopodnra sp. from stalagmite, the surface of
pools and from mud banks and Arrhopalites sp. from
guano, although it may also be an exotic introduction
(Greenslade in lift.). Adelphoderia sp. has been taken
from stalagmite, the surface of pools, mud banks,
flowstone, fungi, guano (1 record) and pitfall (one
record). It was first collected by Hamilton-Smith
around 1964 and was still present in 1988 surveys
despite living in areas which are regularly cleaned
and subject to high tourist visitation. It may be
parthenogenetic as no males have been collected.

Greenslade considered the troglobitic species as
the most important from a conservation standpoint and
the collection sites of most importance as Mammoth,
Orient and Imperial Caves (albeit probably biased by
relative collecting effort).

INSECTA

Specimens belonging to the Blattodea, Orthoptera,
Diptera, Lepidoptera, Hymenoptera and Psocoptera
were deposited in the Entomology collections of the
Australian Museum, however they do not appear to
have been registered in the museum data base.

Order Coleoptera

At least seven beetle taxa belonging to four
families were collected from caves. The carabid
beetles were examined by Dr Barry Moore (then

CSIRO) who identified three species, the most
common being Trechimorphus diemenensis (Bates,
1878) (Fig. 4a). This species is widespread in
southeast Australia, however cave forms possess
shorter wings than surface forms (Moore 1964).
The second species ( Meonis convexus Sloane,
1900) has also been found in the nearby Tuglow
Caves and is possibly troglophilic. The third species
Prosopogmus namoyensis Sloane, 1 895 is considered
to be accidental. The pselaphid beetle Tyromorphus
speciosus (King, 1865) was recorded by Hamilton-
Smith (1966) from the Southern Limestone at Jenolan
(and from caves in Victoria and Queensland). Several
other pselaphids were collected by Gibian et al.
(1988) which probably belong to this species, but this
has not yet been confirmed. The introduced ptinine
‘ spider’ beetle Ptinus exulans Erichson, 1 842 has been
reported from Jenolan (Hamilton-Smith 1967) as well
as many other caves in most Australian states. The
staphylinid beetle Myotyphlus jansoni (Matthews,
1878) was also reported by Hamilton-Smith (1967)
in association with bat guano.

Other unidentified beetles or their larvae have
been collected in Imperial, Mammoth, McKeowns
Hole, Devil’s Coach House and Hennings Cave.

Order Orthoptera

Cave crickets ( Cavernotettix sp.) are commonly
encountered trogloxenes in the entrance, twilight and
transition zones of caves (Fig. 3a). The species from
Jenolan is closely related to those from other karsts in
the region but remains undescribed.

Order Hemiptera

Dr Lionel Hill examined the material collected,
noting some root feeding Coccoidea, one lygaeid
nymph and two species of the dipsocoroid genus
Ceratocombus. One may be C. australiensis Gross,
1950 but the other is undescribed. Both also occur
on leaf litter in epigean habitats and are therefore
regarded as troglophiles.

Order Diptera

Diptera collected or reported include sciarids
(Chaetosciara sp. and Corynoptera sp.), tipulids and
chironomids. They have not been identified and all
are considered to be accidental or trogloxenes.

Order Lepidoptera

The guanophilic tineid moths Monopis
crocicapitella (Clemens, 1859) and Hofmannophila
pseudosprettella (Stainton, 1849) have been reported
from within the caves associated with bat guano.
Hamilton-Smith (1967) reported that Monopis sp.

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

moths have been “found in almost all bat-inhabited
caves of eastern Australia, where the larvae develop
on heaps of guano”. Both species of moth are
cosmopolitan.

Order Hymenoptera

Ants collected in Hennings and Mammoth Caves
remain unidentified.

Order Psocoptera

Booklice have been collected from detritus and
guano in Mammoth, McKeowns Hole and Arch Caves.
One cosmopolitan psocid (Psyllipsocus ramburii
Selys-Longchamps, 1872) has been reported from
many caves in Australia (Smithers 1964) as well as
other situations and is considered to be a troglophile.
The Jenolan material has not yet been identified.

ARACHNIDA

Order Acarina

Numerous mites belonging to four orders were
collected both in and around the caves at Jenolan by
Gibian et al., Eberhard and Holm. At least twenty-
three mesostigmatid taxa (including Uropodina)
were collected within the caves either in leaf litter
accumulations or bat guano. Dr Bruce Halliday
(2001) has published his findings on the Jenolan
Mesostigmata (excluding Uropodina) and has
provided preliminary information on a paper currently
in preparation on the Uropodina. A single parasitic
tick extracted from guano in Paradox Cave has been
identified (considered to be a reptile parasite) but the
remaining Jenolan mite fauna has not been further
examined. This includes mites from three families
of the suborder Prostigmata found in low numbers
in various caves. Oribatids were present in most
samples and were sometimes abundant. No work has
been done on these two suborders at Jenolan.

Most of the mites collected are also known from
surface habitats. Four species have been described
from the Jenolan cave material (Halliday 2001) and
descriptions of a further four Uropodina species
are pending (Halliday in litt. 2013). None of the
mites described displayed morphology associated
with adaptations to subterranean life. We have
tentatively classified about half of the recorded taxa
as troglophiles on the basis of their being recorded,
to date, only from within the caves or having been
recorded in caves on several occasions, even though
some are also well known from surface habitats.

Order Araneae

Spiders are the most commonly seen arachnids in
surface and cave habitats at Jenolan. The best known

species is the troglophilic ‘social spider’, Badumna
socialis (Rainbow, 1905) (Desidae, Fig. 3b), whose
sheet webs are common on the roof and walls of
Jenolan’s Grand Arch through which the road passes.
Their web density can be so great that individual
webs merge to fonn a single large sheet, punctured
by the entrance holes of each spider. Clumps of web
periodically fall off the roof, and it was suggested
that dust and chemical pollution from vehicles might
be adversely affecting the population (James et al.
1990). While it was found that the webs were highly
polluted by lead from vehicle exhaust fumes (Hose
et al. 2002), direct effects on the spider population
were not demonstrated, but continuing monitoring of
the arch population was recommended. The species
is also found in arch habitats at Colong, Abercrombie
and Wombeyan. Few are seen in caves beyond the
cave arch and entrance regions, where local air
currents (and night lighting) probably bring in a
steady supply of insect food. The genetic relationships
between the different arch populations, and a close
surface relative, Badumna Jonginqua (Koch, 1867)
need testing to properly assess their taxonomic and
conservation status. A limited protein electrophoretic
study (Gray, unpublished) showed phylogeographic
differentiation between the Jenolan and Wombeyan
populations. Stiphidion facetum Simon, 1902, a
widely distributed surface species, is also commonly
seen in hammock-like sheet webs on the walls of the
Grand Arch (Figs 3c and 3d).

The first spider described from Jenolan Caves
was the troglophilic linyphiid, Laetesia weburdi
named for the Head Guide, Joseph Wiburd (name
misspelt by Urquhart). Laetesia weburdi (Fig. 3e) is a
relatively small spider with slender legs and variable
pigmentation (dark to pale). It is found in small sheet
webs suspended from walls and formation. The
species was originally placed in genus Linyphia, but
in reassigning this species to Laetesia, van Helsdingen
(1972) noted its close similarity to species from cave
and surface habitats in south west Australia, notably,
L. mollita Simon, 1908 (the type species of the
genus). A second linyphiid, as yet undescribed, is a
troglobitic species, lacking both pigment and eyes.
It is smaller and much rarer than L. weburdi and is
known only from one male (in poor condition) and
juveniles. The webs are similar to those of L. weburdi
and were associated with moist formation in Imperial
and River Caves. Recent searching has so far failed
to find the additional material necessary to properly
describe the species.

An interesting group of theridiid spiders are
tentatively placed in the genus Icona , otherwise only
known from the subantarctic islands of New Zealand
(Forster 1955a and 1964). They were originally

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

placed in Steatoda (in Gray 1973), and subsequently
reassigned to ‘ in or near’ Icona following examination
by H.W. Levi (pers. comm.). These spiders, currently
undescribed, are distributed across southern Australia
as troglophilic and troglobitic species. At Jenolan
there is at least one unidentified species of this group
(Fig. 4d), a relatively common troglophile with
varying degrees of depigmentation (it was at first
thought to represent two species). These troglophiles
were described as having “scatty webs over mud or
leaf litter deposits” (Gibian et al. 1988).

A small troglobitic species of theridiid from Hennings
Cave (Fig. 4e) was also previously included under
Icona (in Gibian et al. 1988, Eberhard and Spate
1995). The taxonomic placement of this eyeless and
totally depigmented species must wait until adult
specimens are available.

A troglophilic species of Cryptachaea is widely
distributed in south-eastern Australia: C. gigantipes
(Keyserling, 1890) is recorded from a number of
NSW caves (Smith et al. 2012), including Jenolan
(previously as Achaearanea veruculata (Urquhart,
1885) in Gibian et al. 1988, Eberhard and Spate
1995). This large species makes a typical theridiid
‘gumfoot’ capture web.

Like the linyphiids, the other web building
troglophiles are very small. These include
members of two surface litter dwelling families:
Micropholcommatidae (0.5-1. 5 mm) spiders found
on mud banks, in litter debris and in small webs on
stalagmites; Mysmenidae (up to 2 mm) where a male
was taken from a ‘small web’. On close examination
these webs are usually seen to be modified orb webs.

Small theridiosomatid spiders ( Baalzebub sp.)
are often seen in cave entrance, twilight and transition
zones in their distinctive cone-shaped orb webs. These
spiders use a central tension line to maintain this web
shape; they release the tension when prey approaches
allowing the sticky orb web to rebound over it.

The non web-building spider fauna includes
several vagrant troglophilic hunters often associated
with loose rock, soil bank, guano deposit, litter
detritus and root mass habitats. Most belong to genera
endemic to Australia and New Zealand. Cycloctenus
abyssinus (family Cycloctenidae) has been
periodically recorded in cave habitats. The original
description (by Urquhart 1 890) was of a female and
subadult males; Rainbow (1893) described an adult
male and referred to several female specimens, but
the whereabouts of these are unknown, and today
there are no pre-1900 specimens or males currently
recorded in the Australian Museum collections.
These spiders are well pigmented and have large eyes

and are probably conspecific with a surface species.
The spiders are not often seen, but are probably an
important predator in the caves ecosystem.

Kaiya terama Gray, 1987 (Gradungulidae) has
been found in several caves at Jenolan, and is a
common epigean log and litter dwelling species.

Tasmanoonops spp. (Orsolobidae) are much
smaller spiders that are found in similar surface
habitats. They have been collected in Elder and
Hennings caves associated with moist habitats,
including hanging root masses.

Order Opiliones

One troglophilic species, the triaenonychid
Holommcia cavernicola (Fig. 3f) was originally
described from “Jenolan Caves” (Forster 1955b) and
re-described by Hunt (1992) based on the holotype
and additional material collected by Gibian et al.
(1988), Hunt and others. While the species regularly
occurs in caves at Jenolan, specimens are also found
in epigean habitats. The harvestman in caves at
Tuglow is tentatively assigned to H. cavernicola.
Other species in the genus Holommcia are found
within multiple karsts in southern New South Wales.
Pigmentation and eye size varied between cave and
surface populations but also within cave populations
(Hunt 1992).

A second species of harvestman, the neopilionid
Megalopsalis sp. is known from two specimens
collected from the entrance chamber of Mammoth
Cave and is probably accidental in caves.

Order Pseudoscorpiones

Three species have been collected at Jenolan.
One is probably an accidental; the other two were
described by Chamberlin (1962) with only vague
locality data but have since been confirmed to occur
at Jenolan. Sathrochthonius tuena (Fig. 4b) is a
guanophile from Bow and Paradox Caves as well
as from Wombeyan Caves. The other is a troglobite,
Pseudotyrannochthonius jonesi (Fig. 4c) known from
Imperial Cave and the Chevalier extension.

MYRIAPODA

Order Geophilomorpha

A geophilomorph centipede seen on flowstone in
Hennings may be an accidental.

Order Polydesmida

Polydesmid millipedes collected from several
caves are considered to be troglophiles. No further
work has been carried out.

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

CRUSTACEA

Gibian et al. (1988) recorded the first aquatic
cave fauna from Jenolan, reporting amphipods
(Crangonyctidae), copepods (Harpacticoida,
Cyclopoida) and ostracods. This material, augmented
by the more extensive collections of Eberhard (1993),
has been re-examined and some identifications
amended to at least six aquatic taxa.

Order Cyclopoida

At least two, possibly three, species of copepod
have been collected in Mammoth and Lucas caves.
The two species that have been identified are well
known surface copepods and may be accidentals
or stygophiles. The third putative species remains
unidentified.

Order Isopoda

Two species of terrestrial oniscid slaters have
been collected, one strongly pigmented and eyed from
Elder Cave, the other is a single weakly pigmented
specimen (, Styloniscus sp.) from Mammoth Cave. We
have been unable to locate the Elder Cave specimen
and the Styloniscus specimen has not been further
studied.

One species of aquatic phreatoicoid isopod
(Crenoicus sp.) has been netted in both the Imperial
resurgence and in Paradox Cave by Eberhard. It is
likely stygophilic but has not been further studied.

Order Amphipoda

Eberhard trapped the eusirid amphipod
Pseudomoera fontana (Sayce, 1902) in both the
Northern Stream sink and the Imperial Cave
resurgence; it is a common species in southeast
Australian streams and is either an accidental or
stygophile.

Neoniphargid amphipods were trapped in both
Paradox Cave and the Imperial streamway. Bradbury
and Williams (1997) described the stygobitic
Neocrypta simoni based on the material collected
by Stefan Eberhard in Paradox Cave (Fig. 4f); five
specimens netted in the Imperial River by Gibian,
Smith and Wheeler have not been identified as yet.

Order Anaspidacea

Eberhard (1993) collected stygobitic syncarids
(Psammaspididae) by placing baits (kippers in
brine) in the Imperial and Spider Cave rivers and
in perched seepage fed pools well above the river
level. Mia Thurgate collected more from the Pool of
Reflections in River Cave in 2000. Psammaspidids

(Fig. 4g) are a primitive group of eyeless crustaceans
recorded from ground waters in eastern Australia. No
further taxonomic work has been conducted on this
interesting material.

MOLLUSCA

Class Gastropoda

Pommerhelix depressa (Hedley, 1901) and
Elsothera sericatula (Pfeiffer, 1849) have been
collected in Casteret Cave and caves in the southern
limestone. Eberhard collected the aquatic snail
Glacidorbis hedleyi Iredale, 1943 at the Imperial
resurgence. Snails collected by Gibian et al. (1988)
have not been examined.

ANNELIDA

Terrestrial and aquatic oligochaetes were reported
by Gibian et al. (1988) and Eberhard (1993) but not
further identified.

NEMATODA

Terrestrial and aquatic nematodes were reported
by Gibian et al. (1988) and Eberhard (1993) but not
further identified.

PLATYHELMINTHES

Flatworms of the Orders Paludicola and Terricola
were reported by Eberhard from Wiburds Lake,
Mammoth and Serpentine Caves.

DISCUSSION

Comparisons of biodiversity patterns between
different karst areas can be fraught with biases
including, inter alia, area effects and differences
in survey effort, methods and taxonomic biases, as
well as bias towards troglobitic/stygobitic species,
incorrect ecological classification, provincialism and
other fallacies (see Culver et al. 2013). Nevertheless
we consider it timely to undertake a brief re-appraisal
of Jenolan’s cave fauna to place its significance in a
regional and national context, especially because a
great deal of subterranean fauna research has occurred
elsewhere in Australia (see Guzik et al. 2011) since
the previous Jenolan and New South Wales inventory
by Eberhard and Spate (1995); Thurgate et al. (2001a,
2001b).

Jenolan retains its status with the highest

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

recorded subterranean taxonomic diversity (136
taxa) of any karst area in New South Wales, which
is at least partly an artefact of high survey effort,
with Jenolan drawing the attention of biologists over
many decades. Notwithstanding, we hypothesise
that other environmental factors may be responsible.
Firstly, Jenolan is highly karstified and hosts the most
extensive cave system in New South Wales with over
40km of surveyed cave passage. This subsurface
‘area effect’ is consistent with studies elsewhere (e.g.
Graening et al. 2006) which show increasing cave
length correlates with increasing species richness.
Secondly, Jenolan is a topographically diverse fluvial
karst with many large-sized cave entrances (vertical
and horizontal) and multiple sinking streams which
facilitate active colonisation of caves by animals, but
also particularly, their passive transport underground
(by gravity, water or air), which may partly account
for the high proportion (53%) of taxa classified as
‘cave accidentals’ in our inventory. This ratio is not
dissimilar to 42% recorded in a desktop bio-inventory
of the Nullarbor which is a significantly larger
karst area (by > 2 orders magnitude) but similarly
characterised by multiple large-sized cave entrances
where collecting efforts have historically tended to
focus (Eberhard in lift.).

While the classification of taxa as ‘accidentals’
or otherwise (trogloxene, troglophile, troglobite)
is often necessarily inferred owing to limitations in
survey data and knowledge of species taxonomy and
ecology, ambiguous classification or misinterpretation
of troglomorphic traits may skew interpretation of
site ‘significance’ when assessed in terms of total
species richness. For this reason, many comparisons
between karst areas in the literature are restricted
(arguably biased) towards troglomorphic species
(presumed troglobites and stygobites, see Culver
and Sket (2000). Notwithstanding, troglobites and
stygobites are more typically short-range endemic
species and therefore more vulnerable to threats and
extinction from environmental changes. On this basis
a high conservation significance may be attributed to
troglobites and stygobites.

In paving the way for standardized and
comparable subterranean biodiversity studies, Culver
et al. (2013) concluded that it is necessary to treat
troglobites and stygobites differently from non-
obligate species, because differences of opinion exist
as to which species are troglobites and stygobites. In
our opinion the eight species considered likely to be
troglobites or stygobites at Jenolan (revised from 14
troglomorphic species earlier reported by Thurgate et
al. 2001a) ranks as fairly typical for karst areas in the
eastern highlands (Eberhard and Spate 1995). At this

point in discussion it is appropriate to correct an error
in the Jenolan Karst Conservation Reserve Draft Plan
of Management (Department of Environment and
Conservation NSW, undated, p. 49) which mistakenly
reports 147 species of troglobitic [sic] fauna.

We consider it likely that additional obligate
subterranean species remain to be discovered at
Jenolan, especially in the poorly sampled epikarst,
vadose, deep phreatic and interstitial aquatic habitats,
and terrestrial meso-cavern habitats. Our prediction is
based partly on the diversity known from Wombeyan
Caves, located 55 kilometres south of Jenolan,
which has a high diversity (11 species) of stygobitic
amphipods (Bradbury and Williams 1997). For
comparison, the richest obligate cave fauna recorded
from eastern Australia is Bayliss Cave, a lava tube
in north Queensland, with 20 species of troglobites
(Culver and Sket 2000). Tasmania is also relatively
diverse with 15 or more obligate species recorded
from well-developed karst areas (Eberhard 1996).

The fallacy of provincialism as termed by
Culver et al. (2013) occurs when data from one
‘favoured’ place is treated differently than data from
other places. In applying the metaphor ‘from rags
to riches’ to highlight subterranean biodiversity in
New South Wales, Thurgate et al. (2001a) may have
been justifiably optimistic, however, this paradigm
deserves to be reappraised in the national context
considering subsequent discoveries of remarkably
diverse subterranean faunas in other states. Recently
in Western Australia sampling of deep groundwater
aquifers has revealed the existence of diverse (> 60
species) stygobite communities (e.g. Eberhard et al.
2009). Sampling of terrestrial meso-cavern habitats
in iron-ore and calcrete rocks has also revealed highly
diverse troglobite communities comprising > 45
obligate species (S. Eberhard in lift ).

FUTURE RESEARCH AND CONSERVATION

PRIORITIES

The Jenolan Karst Conservation Reserve Draft
Plan of Management (Department of Environment
and Conservation NSW, undated) recognises that
cave fauna is highly susceptible to disturbance and
recommends further investigation into the potential
impacts of human activities on the conservation
of these species. The material from the 1986-1993
collections represent a reasonable baseline survey for
Jenolan Caves. Nevertheless cave fauna, especially
the highly adapted species, are usually rare and it is
highly likely that further intensive collection efforts
would result in new taxa being found. Alternative

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

collection techniques used for aquatic micro-fauna
and terrestrial meso-cavern habitats e.g., damp leaf
litter packs (Weinstein and Slaney, 1995) should be
evaluated as they may effectively sample taxa that
were not collected using the methods previously
employed. The current state of knowledge, gaps and
research priorities are summarised in Table 2.

A great deal of the material collected has not yet
been sorted to species level. New species still await
formal description due to the very limited funding
and diminishing taxonomic resources available in
Australia. Future collection efforts could concentrate
on obtaining specimens of groups where a funded
taxonomist is available, or aim to increase the
number and quality of specimens of certain important
troglobitic and stygobitic representatives (e.g. by
obtaining more mature material, including both sexes)
or seek information on their biology and ecology,
about which virtually nothing is known.

The species of most conservation interest are those
species restricted to the subterranean environment,
especially the troglobites and stygobites. The physical
extent and degree of karstification at Jenolan, and the
hypothesised presence of undiscovered troglobitic and
stygobitic taxa in the mesocavern and other cryptic
aquatic habitats, emphasises the importance of the
continuing biological exploration of this significant
subterranean ecosystem.

ACKNOWLEDGEMENTS

We would like to thank Louise Smith (previously L.
Wheeler), the Jenolan Caves Reserve Trust and the Australian
Museum for their support with the 1986-93 collection work
and especially Ernest Holland for his supervision, support
and advice. We are indebted to the following taxonomists
who have worked on the Jenolan fauna: Dr Chris Allen, Dr
Max Beier, Dr John H. Bradbury, Dr Cathy Car, Dr Peter
Cranston, Dr Alison Green, Dr Penelope Greenslade, Dr
Bruce Halliday, Mr Danilo Harms, Dr Mark Harvey, Dr
Lionel Hill, Dr Glenn Hunt, Dr Tomislav Karanovic, Dr
Robert Mesibov, Mr Graham Milledge, Dr Barry Moore,
Dr Ebbe Nielsen, Dr Winston Ponder, Mr Peter Serov, Dr
John Stanisic, Dr Michael Rix, Professor William (Bill)
Williams, Dr George (Buz) Wilson.

REFERENCES

pdf files of unpublished reports denoted with asterix(*) are
available from the author for correspondence

Barnard, J.L. and Karaman, G.S. (1982). Classificatory
revisions in gammaridean Amphipoda (Crustacea),
Part 2. Proceedings of the Biological Society of

Washington 95, 167-187.

Beier, M. (1966). On the Pseudoscorpionidea of Australia.
Australian Journal of Zoology 14 , 275-303.

Beier, M. (1967). Some Pseudoscorpionidea from
Australia, chiefly from caves. The Australian
Zoologist 14 , 199-205.

Bradbury, J. H. and Williams, W. D. (1997). The amphipod
(Crustacea) stygofauna of Australia: description of
new taxa (Melitidae, Neoniphargidae, Paramelitidae),
and a synopsis of known species. Records of the
Australian Museum 49 , 249-341.

Chamberlin, J. (1962). New and little known false

scorpions, principally from caves, belonging to the
families Chthoniidae and Neobisiidae. Bulletin of the
American Museum of Natural History 123 , 301-352.

Chen, J., Leng, Z, and Greenslade, P. (2005). Australian
species of Sinella ( Sinella ) Brook (Collembola:
Entomobryidae). Australian Journal of Entomology
44 , 15-21.

Clark, S.A. (2009). A review of the land snail genus
Meridolum (Gastropoda: Camaenidae) from central
New South Wales, Australia. Molluscan Research 29 ,
61-120.

Culver, D.C. and Sket, B. (2000). Hotspots of subterranean
biodiversity in caves and wells. Journal of Cave and
Karst Studies 62 , 11 - 17 .

Culver, D.C., Trontelj, P, Zagmajster, M., and Pipan,

T. (2013). Paving the way for standardized and
comparable subterranean biodiversity studies.
Subterranean Biology 10 , 43-50.

Dew, B. (1963). Cave animals. Journal of the Sydney
University Speleological Society 6 , 10 - 28 .

*Eberhard, S. (1993). Survey of fauna and human impacts
in the Jenolan Caves Reserve. Unpublished report
produced for the Jenolan Caves Reserve Tmst, New
South Wales, Australia. April 1993. 16pp.

Eberhard, S.M. (1996). Tasmanian cave fauna. In
‘Encyclopedia Biospeologica Tome III.’ (Eds. C.
Juberthie and V. Decu) pp. 2093-2103. (Societe
Internationale de Biospeleologie, Moulis- Bucarest).

*Eberhard, S. and Spate, A. (1995). Cave invertebrate
survey: Toward an atlas of NSW Cave Fauna. 112pp.
(A report prepared under the New South Wales
Heritage Assistance Program NEP 94 765. November
1995).

Eberhard, S.M., Halse, S.A., Williams, M., Scanlon, M.D.,
Cocking, J.S. and Barron, H.J. (2009). Exploring the
relationship between sampling efficiency and short
range endemism for groundwater fauna in the Pilbara
region, Western Australia. Freshwater Biology 54 ,
885-901.

Forster, R. R. (1955a). Spiders from the subantarctic
islands of New Zealand. Records of the Dominion
Museum Wellington 2 , 167-203.

Forster, R.R. (1955b). Further Australian harvestmen
(Arachnida: Opiliones). Australian Journal of
Zoology 3 , 3 54-4 1 1 .

Forster, R. R. (1964). The Araneae and Opiliones of the
subantarctic islands of New Zealand. Pacific Insects
Monograph 7, 58-115.

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

Table 2. State of knowledge, gaps and research priorities.

Higher Group

Relative

Diversity

Taxonomic

Resolution

Comments

Research Priorities

Entognatha:

Collembola

High

Good

Well sampled and
identified, includes
troglobites and
undescribed n. sp.

Describe n. sp. especially
troglobites

Insecta:

Coleoptera

Moderate

Good

Well sampled and
identified in macro-cavern
habitats but meso-cavern
habitats poorly sampled

Sample meso-cavern
habitats

Insecta: Others

Moderate

Poor

Poor taxonomic resolution

Identify existing collections

Arachnida:

Araneae

High

Good

Generally well sampled
and identified, but includes
rare troglobites and
undescribed n. sp.

Targeted sampling of
troglobites and describe n.
sp.

Arachnida:

Acarina

High

Good

Well sampled and
identified

Describe n. sp.

Arachnida:

Others

Low

Good

Well sampled and
identified in macro-cavern
habitats but meso-cavern
habitats poorly sampled

Sample meso-cavern
habitats

Crustacea

Moderate

Poorly sampled, likely
to be more diverse,
especially micro-crustacea

Sample deep aquatic
habitats, identify and
describe n. sp.

Myriapoda

Low

Poor

Poor taxonomic resolution

Identify existing collections

Gastropoda

Low

Excellent

Terrestrial snails sampled
and identified, aquatic
snails poorly sampled
(Hydrobiidae)

Sample deep aquatic habitats

Annelida,

Nematoda,

Platyhelminthes

Low

Poor

Poorly sampled, likely
to be more diverse,
especially aquatic
Oligochaeata

Sample deep aquatic habitats

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

Forster, R.R., Platnick, NT. and Gray, M R. (1987). A

review of the spider superfamilies Hypochiloidea and
Austrochiloidea (Araneae, Araneomorphae). Bulletin
of the American Museum of Natural History 185 ,
1-116.

*Gibian, M., Smith, G. and Wheeler, L. (1988). Interim
report on the survey of the invertebrate fauna of
Jenolan Caves 3 rd August, 1988. 12 pp. Unpublished.

Graening, G. O., Slay, M.E. and Bitting, C. (2006). Cave
fauna of the Buffalo National River. Journal of Cave
and Karst Studies 68 ( 3 ), 1 53-163.

Gray, M. (1973). Survey of the spider fauna of Australian
Caves. Helictite 11 , 47-75.

Gray, M.R. (1983). The taxonomy of the semi-communal
spiders commonly referred to the species Ixeuticus
candidus (L. Koch) with notes on the genera
Phryganoporus, Ixeuticus and Badumna (Araneae,
Amaurobioidea). Proceedings of the Linnean Society
of New South Wales 106 , 247-261.

Greenslade, Penelope (1992). The identity of Australian
specimens recorded as Lepidosinella armata
Handschin 1920 (Collembola: Entomobryidae) with a
key to Australian Sinella and Coecobrya. Journal of
the Australian Entomological Society 31 , 327-330.

Greenslade, Penelope (2002). Systematic composition and
distribution of Australian cave collembolan faunas
with notes on exotic taxa. Helictite 38, 11-15.

^Greenslade, Penelope (2011). Collembola from Jenolan
caves. Unpublished report to the New South Wales
National Parks Service. November 1989, amended
July 1993, corrected May 2011.

Guzik, M., Austin, A., Cooper, S., Harvey, M.,

Humphreys, W., Bradford, T., Eberhard, S.M., King,
R., Leys, R., Muirhead, K., Tomlinson, M. (2011). Is
the Australian subterranean fauna uniquely diverse?
Invertebrate Systematics 24 ( 5 ), 407-4 1 8. http://
dx.doi.org/10.1071/IS 10038.

Halliday, R.B. (2001). Mesostigmatid mite fauna
of Jenolan Caves, New South Wales (Acari:
Mesostigmata). Australian Journal of Entomology
40 , 299-311.

Halliday, B. and Masan, P. (2008). Pachydellus hades
(Halliday) (Acari: Pachylaelapidae), a European mite
species described from Australia. Australian Journal
of Entomology 47, 225-226.

Hamilton- Smith, E. (1966). Pselaphidae from Australian
Caves. Journal of the Entomological Society of
Queensland 5, 70-71.

Hamilton- Smith, E. (1967). The Arthropoda of Australian
Caves. Journal of the Australian Entomological
Society 6, 103-118.

Hamilton-Smith, E. and Adams, D.J.H. (1966). The

alleged obligate ectoparasitism of Myotyphlus jansoni
(Matthews) (Coleoptera: Staphylinidae). Journal of
the Australian Entomological Society 5 , 44 — 45.

Harms, D. and Harvey, M.S. (2009). A review of the
pirate spiders of Tasmania (Arachnida, Mimetidae,
Austral omimetus) with description of a new species.
Journal of Arachnology 37, 188-205.

Harms, D. and Harvey, M.S. (2013). Review of the cave-
dwelling species of Pseudotyrannochthonius Beier
(Arachnida: Pseudoscorpiones: Pseudotyrannochth-
oniidae) from mainland Australia, with description
of two troglobitic species. Australian Journal of
Entomology 52 , 129-143.

Heimer, S. (1986). Notes on the spider family

Mimetidae with description of a new genus from
Australia (Arachnida, Araneae). Entomologische
Abhandlungen. Staatliches Museum fiir Tierkunde
Dresden 49, 113-137.

Helsdingen, PJ. van (1972). An account of money

spiders from down under (Araneida, Linyphiidae).
Zoologische Mededelingen 47 , 369-390.

Hose, G.C., James, J.M. and Gray, M.R. (2002). Spider
webs as environmental indicators. Environmental
Pollution 120 , 725-733.

Hunt, G.S. (1992). Revision of the genus Holonuncia
Forster (Arachnida: Opiliones: Triaenonychidae)
with description of cavernicolous and epigean species
from Eastern Australia. Records of the Australian
Museum 44 , 135-163.

James, J.M., Gray, M. and Newhouse, D.J. (1990). A
preliminary study of lead in cave spider’s webs.
Helictite 28 , 37 — 40.

Moore, B. P. (1964). Present day cave beetle fauna in
Australia: a pointer to past climate change. Helictite
3, 69-74.

Office of Environment & Heritage NSW (October
2013). Draft Plan of Management Jenolan Karst
Conservation Reserve. NSW National Parks and
Wildlife Service. 102 pp. http://www. environment.
nsw.gov.au/parkmanagement/JenolanKCRdraftPOM.
htm

Rainbow, W. J. (1893). Descriptions of some new

Araneidae of New South Wales. No. 1. Proceedings
of the Linnean Society of New South Wales (2) 7,
471-476.

Rainbow, W. J. (1904). Studies in Australian Araneidae HI.
Records of the Australian Museum 5 , 326-336.

Rainbow, W. J. (1905). Studies in Australian Araneidae IV.
Records of the Australian Museum 6, 9-12.

Rainbow, W. J. (1911). A census of Australian Araneidae.
Records of the Australian Museum 9, 107-319.

Richards, A.M. and Lane, E.A., (1966). Exotic Collembola
from Jenolan Caves, N.S.W. Helictite 4 , 88-89.

Rix, M. G. and M. S. Harvey. (2010). The spider family
Micropholcommatidae (Arachnida, Araneae,
Araneoidea): a relimitation and revision at the
generic level. ZooKeys 36 , 1-321.

Smith, H., Vink, C.J., Fitzgerald, B.M. and Sirvid, PJ.
(2012). Redescription and generic placement of
the spider Cryptachaea gigantipes (Keyserling,

1890) (Araneae: Theridiidae) and notes on related
synanthropic species in Australasia. Zootaxa 3507 ,
38-56.

Smithers, C.N. (1964). New records of cave and mine-
dwelling Psocoptera in Australia. Australian Journal
of Entomology 3 , 85.

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

Thurgate, M.E., Gough, J.S., Spate, A. and Eberhard, S.
(2001a). Subterranean biodiversity in New South
Wales: from rags to riches. Records of the Western
Australian Museum Supplement 64 , 37-47.

Thurgate, M.E., Gough, J.S., Clarke, A.K., Serov, P.
and Spate, A. (2001b). Stygofauna diversity and
distribution in Eastern Australian cave and karst
areas. Records of the Western Australian Museum
Supplement 64 , 49-62.

Urquhart, A. T. (1890). On two species of Araneae new
to science, from the Jenolan Caves, New South
Wales. Transactions of the New Zealand Institute 22 ,
236-239.

Weinstein, P. and Slaney, D.P (1995). Invertebrate faunal
survey of Rope Ladder Cave, North Queensland: a
comparative study of sampling methods. Journal of
the Australian Entomological Society 34 , 233-236.

Appendix
(next 18 pages)

Invertebrate fauna collected within the caves at Jenolan

* Specimens identified by Dr C.B. Allen (CA), Dr M. Beier (MB), Dr J.H. Bradbury (JB), Dr C. Car
(CC), Dr P. Cranston, Dr M. Gray (MG), Dr A. Green (AG), Dr P. Greenslade (PG), Dr B. Halliday (BH),
Dr M. Harvey (MH), Dr L. Hill (LH), Dr G. Hunt (GH), Dr T. Karanovic (TK), Dr H.W. Levi (HL), Dr
R. Mesibov (RM), Mr G. Milledge (GM), Dr B. Moore (BM), Dr E. Nielsen (EN), Dr W. Ponder (WP),
Dr M. Rix (MR), Mr P. Serov (PS), Dr H. Smith (HS), Dr J. Stanisic (JS), Prof. W. Williams (WW), Dr
G. Wilson (GW). Typ = type specimen(s). References to Smith as collector are G. Smith unless indicated
otherwise.

** Native or introduced/cosmopolitan

*** Ecological Status: Accidental (Ac), Guanophile (Gp), Stygophile (Sp), Stygobite (Sb), Troglophile
(Tp), Troglobite (Tb), Trogloxene (Tx)

**** Institutional abbreviations: Australian Museum, Sydney (AMS), American Museum of Natural
History, New York (AMNH), Australian National Insect Collection, Canberra (ANIC), National Mu-
seum of New Zealand, Wellington (NMNZ), South Australian Museum, Adelaide (SAMA)

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

C /5

A A

^ o

(D E
E S3

Cd Cd

•S A
O C/2

C /5

A A
8«
CJ o

CD E
E Sd

Cd Cd

U-/ O

•S A
O C/2

C /5

A A

8-2

— -2
CD E

e sd

Cd Cd

U-/ j

M A

O t /2

C /5

A A

8 2

(D E

e sd

Cd Cd

U-/ j

M A

O ! si

C /5

X HH

o A

d o

C /5

C/2

C /5

A A

<N £

C/5

o ^

C/5

—

o
o

w
CD
Td
cd

C/5

CD ■ j

•S A

O C/2

C/5

—

U-/ j

M A

O C/2

C/5

—

C/5

—

U-/ j

M A

O C/2

U-/ j

M A

O C/2

. o

cd C/5

60 sd

b CD

2 o
-O o
H O

C/5

3 d

D E

3 «

E sd

cd cd

D Cd

—

U-/ j

M A

O C/2

% e
2 o
sd 3
cd cd

M ‘G

• - o

<D c /5

'o ^

CD c /5

c /5 e

— H S3
.2 03
J-H C /5 C /5

■S -K &P

CD (D O
H »-H

d £ W>

1 1:1

td td a

cd cd ^

c o Sd

„ =s o

j-r o ._h

O H M

o.S d
£ %<£
2 O eg
“ y >

CD t

Q o

. <*>

-o'

-(->

Oh'O

o o

S o

55 Td

o o

o sd
-(->

"S ® s

.2 CD S

Ch C/5 T

O 3 2

H*' HtH ^

cd i-h

'£ 4d
D O
SO cd

E A o

HH W 1/5

D cd

C/5 1

O
o

Jd
o
cd
O

C/5

+->

C/5

cd ai

2 2

C/5

S— i

•-H 00

^ .2

^ o

O' D
Oh

o ^

C/5
D
.;>
<h

^ C

PQ eg

C/5

c/o'

/-H

C/5

r“

C/5

O-

(L>

C/5

H— »

C/5

C^H

C/5

”2

W
sd C/2

cd _h

,+d .o

O D

so -

C/5

O
O Td
D

C/5

-S

S— <

c 2 .2

D 2

§ cd -E5
'E 2 O
O

C/5

ffi

C/5

+->

C-H

C+H

C/5

2 o

s a

H— »

D C/5 D

Sd Cd <

+-*

D E
O D

cd dr D

' Nj j$

•|*»4

5s.

»5

5s.

s ^

& 2
r 3 6 D

r 2 co

^ =>
•0 D
E: cd
cj O
o. sd

K E

o d

"53d

<0-H

S— <

s—

Oh ^

00 O

q pq

5s.

Co S

£?

C/5

5s.

-a

dQ*

C/5

+~H

"§

6J2

a >n
o rn
8 2°
05

E d

a 3 h

S S

8 3

C O

so t 3

G a*
•fc

Co ^

3 a
^ a

H — *

C/5

3 J

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

C/5

CL)

A Z

— -2
CD T3
T3 O
kS KS

c/5

§ ^
O t/2

C/5

A Zj

8 <
^ o
(D T3
T3 O

C3 c3

c/5

CD j

O t/2

C/5

A Zj

8 «
^ c
(D T3
T3 O
kS c3

c/5

CD j

O t/2

C/5

A Zj

8 <

<N £

C/5

—

C/5

A Zj

8 <

<N £

C/5

A Zj

8 <

<N £

C/5

A Zj

8 <

<N £

C/5

c3) +2

■ Co

.o -t3

i l

■§5

On O
c/5 c

. - o

O C/5
•rt c/5

o §

CD H
Oh,”
R T3
CD C
5/3 c3

T3
03
R

c3
Oh

cd"

CD R

5/3 to

r, CD

CD H
C/5 JrH
r-f CD

E >

C/5

O
c3

c3
> X)

C/5

S-

H->

CD"

J-*

i-H

5/3

'-O

C/5

„ W

Oh 5/3

Ch-H

C/5

‘I

-O

C/5

Ch-H

C/5

S-H

5/3

CD"

C/5

w pq

O ” c3
3 ^ Oh
L o ^

o ■

<D "

D C3
hP

C/5

O o

, Q a

S-H CD ^

0^-0 0

a u pq

to co

CD O

SP o -i

.s s

O 3

H H— I
&

c/5 _D

.2 -c

O <D
CD C/5

c/5 ^3

C/5

-y

H->

O c/5
CD H->

33 §
c O.

o o

C 3 " Oh
CD T-s

fS-H a
Dh Ch

C/P

C/5

CD C/5
> <D

3 c§

^ £

O y

pq 3

'-O

p— *

VO
VD
Q OV

"§

&— I

C/5

CD
O

hP »-!
3 D
T3 +3

ex, o P 3

5T c3 ^
^ M (N

C/5 “ <05

R“

►R

p-i

C/5

!-h O

P— * n

3 hS d

C/5

■ti d
T 3 >;

t 3 a

c3 CJ

T3 +3
c2 Ch

c3 .a

5 ) 3 )

•C £

O Oh

h— v <3)

ri R
8 ^
U %

CD c/5

T3 .D
'o

O Oh
CD t«
D cu

o a

. C/5

£2 T3

^ D

"5 s £
h- 2
^ 3
o o

<o t 3
^ 2
0 3

'-O

-(->

1
£
£

CO-

C/5

CL>

C/5

►R

-a

C/5

53 _

b-?3

5J
CL>

C/5

-P

t-5

R
=0
S

R «

?S 5 h

R «->

O
R

^ t/2

-a

C/5

^R.

c/5

P-H

■£P

— .

C/3

-+R4

Proc. Linn. Soc. N.S.W., 136 , 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

C/5

CL)

U-/

O t/2

C/5

—

U-/

O t/2

C/5

—

U-/

O t/2

a3 ^

§ "§

rS "§
O Oh

C3 ^

H— »

<L>

2 "§
a •§
o *

* -§

c E

•A J

E 'I

"2 GO

O ^

• ^ r — .

>-l fa <

O J

So u .S

00 g TO

O g CD

w 0

.s 00
^ 3 §<

.S Oi ^

£ oo s

O £ £

rc go

r s

CCS <

o 3

ccS ccS

-O

c c/5

S O

S Oh

CD D
Oh X>

C/5

05 £

oo r

05 ^

2 E

— H-l

.B -o

S GO

g 2
3 <

C/5

1
s

£<s

a w

Gh GO

o o

CD -O
Oh c/5
C/5 D

CD c/5
^ C/5

c3 3

O -

C/5

TO
O
o ccS

S E

Oh • —

o 2

O -O

o
4— »
O

rt 3
1 *

C/5

"c3

C/5

L-

-O

C/5

<D ^H
^ O

= S

"3 "3

[-0 GO

O >5

HH X)

'-O

H— »

-O

_0D

"Hb

H-l

C/5

HH.

5h.

§<

~ £$

&h lo

2s KJ

L) ^

C/5

§?

ex,

G-h'

■—

-a

■—

C/5

-a

■3

i/i

■+«»i

H-*

—

H — *

5s.

S-

-a

— i

S-

■—

-a

■—

H->

6JD

— 1

c3
o Oh
3

O hO

E S

-HH

■—

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

+->

to
CL)

bn
bn

O CD .JO

M X) S 3

c 3

s ^
B

CL)

Oh CO
co a
CD *0
co S

-fcs.

5 s.

+->

CL)

i-H

■

^ a 3

| £
£ «=

o jx

p n
O c 3

o -s

CD O

1—5

_. Oh
' O

cd x

5/3

r s

cc! <|
o> ^
§ "§

rS "§
O Os

5/3

r s

cc! <|

§ "§

rS "§

O Oh

5/3

r s

cc! <

+->

CD X

O x

ccS ccS
X
O

Dh
Oh

<u £
bn ffi

Cu - -s

Oh -2
■ <
T 3 rjJ
CD

Oh ("'■

O oo

.22 O

X-»

.22 O

+->

-§ 5

+->

-§ 5

Oh <!

5/3

r s

ccS <|

§ "§
rS "§

O Oh

C/3

r s

ccS <|
o> ^
§ "§
rS "§

O Oh

CD X
CCS CD

h. c 3

ccS O

X O

.s-h

ffi

o ■
Oh

o ■

c- ■
o

o ■

ffi

«o

-O

-tsi

5 s.

•*->

C/3

-O

+->

-O

®S CD

g .B

"s S

Oh C
CD

+-*

H ^

+->

^ V,

-s o

s ^

’Z

C /3

-tsi

‘G

H->

0 D

ffl

-o

0 D

-o

C 5

c 3

+->

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

_CD

‘G

*-h

C/2

„ „ Gh

1 e ^

3 3 O

ffi 3 G

‘I

G, c/2

r s

G <
o> ^
§ G

3 "§

O Gh

CL)

Gh g

<*. co"

C/2 co

-Is

r- ci

g °

C/2

+-*

_CD

‘G

C/2

c/2 o

S> 20
^ 02

co" c/2

id ^

ri - 'n
<N On
- CO 02

2 — /

s-T

■ — 1

20 "

C/5

C/2

02 "

H->

J-H

-c->

k >

C/2

-rt Cl
3 Cl
G Cl

-G o

00 [t]

-G

C/2

G *3

3 ^

n 1 .s

G CD

S '"* 3

c/2 G

£ ^

. §
^ 3 d

— H Ctf

8 S

-T

C/2

in"

C/2

G" 02

c~ o
Cl o
oo >n
G co
c/2 c/2

+->

C/2

L-

C/2

0 > o>

Ch *-H

G G G

S -3 t

ill

W I .2

S G CD

'G C/2 G

O of "§
~ 00 ®
So G ©
CD • G Gh

C 5 G
3 G 3
G (D G

u E £

C/2

ffi

c /2

C/2

G t_i
C/2 (D
w 2

£ w

£
u

CD’ CD

G O
G G

C c /2

S G

Gh ^
S-H O

<d 22

C/2 K

S G
2 c

-B G
G c/2

(D w
c/3 3

oo 3

- e

g |

CD G

E 2

D D

C/2 — '

„ G
G &

£
U

C M

— C/2

Gh CD
> ■ c /2

G -
O CD
G C/2

-, 00 ^

3 £

& s w

Q O G

ffi

cd"

00
G

« c /2

G c«
“ cs

C/2 C2

00 CD

.B 3

G O
G

HtH G

G o
HH O

2 i>

s >

£ °

3 GO

2 -g
^ 0>
X "O

•S 8
s ^

G (_
G O

M— I

02 G

2° Jd

s — ■" 3

£ S

G ^2
G w
co^ C^
D D

C/2
D D

Q 3

00 G
-H O

C/2

o
G
^G

T3 G
D

G T3

•a b
^ g

G 3

o ■
o

o ■

U S
u S

C/2

'O —

"g G

O ^

S 00

g .£

C/2

H->

C/2

h-3

yj j

.Gh

G
^ .

C/3

H-*

—

C/2

^ H

H-*

C/2

-a

— 1

C/3

C/2

H->

—

C/2

-a

*1^4

6-5

H>hH

CO 1

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

C/3

CL)

-E

CL)

-e

CL)

CCS

CL)

-e

C/3

-E

E-

o „

CD C/3

a s>

C/3 w

<D a

-e a
e 2

H ^

. r, 03

^-n CD
°o -g
00 CD
OV X

w 3
cd

-E

e .5

E o

C/3

D OJ

2 £
E E

- 3

C/3

OO <E

2° £

C/3 ^

«n 2
<N ^

00 g?

(-~h Ov

S C/3 ^

C/3

>a>

03
CD

2
= s

1 =

E cd
1/5

OO 3

2? e

2 'C

ri cd

VL/

■d

E B
3 3

+->

<3V

S 3
.2 a

-E

C/3

O X

c—

<3V

-E

c3v

C/3

E _

2 3 s

E «

3 |

° x

-E

o
o

CD
CD

x „

o +->

E 53

3 >

CD a

a c/3

b 3

C/3 e
„ <13
CD X

X
o
E

J» to
CD 'e

Q £

1—1 CD

>; ^

CD T3

CD hR

2 g

O |
C/5 <0

2 S

3 ^
■ B in
&o o
•c

a-*

J-H

G-

f a->

Vh / s

t* ^
>-. 00

3 3

3 w

e 3

CD P
> '
CD

c3v

+->

B ^

o cr

T3 ^

cd R:
fc O

CD “

3 g

C/5

S 03
bo

a w)
00 3

^ o

3 1-1

P E

*T3

2 5 s

0 2

o Ov

2 >>

C/5 cd

2 2
E CD
E CD

j-T

a-*

‘C

E^ ^
E C/3

o 3
§ 2

C/5

o> cd

0J3

C/5

■S o

^ 0
2 ov

00 cd

C/5

to E

E -E

03 t:

to jh a

OD g) (JO

3^ x cs
2 ^

5b e ks

•-H hO O
X <i)

sT“ E

x .a

to ^

QJ ^

2 2
a 3

|v §

C§ O'
2

03 w
(D

Ja >
5a G

2 ^

S 3

E p

2 3

.CD

t-~

<3V

«o

E ■

% 3

•<*»4

/ s

C3V

5s.

C/5

-E

.E 1

C3V

*fe

c3v

’— 1

R 2
r*

bo 00
a ^

5^ .

a cd
R 00
Q 03

’a* • *a

<^)

g £

§ E

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

c /2

c n

/ ’p'

-l-H

i—

-P

^ <N
-Q ri
T3 ^

<2 oo

3 M

CD w

<L>

C/2

CD _
O, • ' Q
C/5 O

-P

C/5

£ _
P 3 ri
a> m

o rl
<d <N
= m

CL)

C/2

«n
P 3 O
(D O

v c /2

o *
o —
-h C/2

■g s
s <

s q

s «

• n _QQ

/ \

00 P

2? 3

OV Ch
CD

— CD

+->

o 6

C/2

-T

C/2

-T

C/2

^ m
02 in
2 D 00

02 -H

ov ri

C/2 C/2

s-T

C/2

-O

C/2

-C->

-Eh

00 .IQ

3 s

C/2 C/2

-P

C/2

T3 ' — ■>
D 2D

o g
3 ^
3 3

O c /2

C/2

1—

H — *

C/2

■ — ^

-C->

Q-*

C/2

H — *

■ — ^

C/2

CD
Q
CD
-P

1/5 —H

CD P3

Qh £

H xi

t-~

f CD

C/2

^ C/2

' ^

P kS

P >-.

P &

-Q

B Si

O 00

_d 02

O ^

0 ^

1 ^

S S
s %

00 a 3

rjs Q)
d>

+->

rS g C/2

O C^ ^

'+H

_op

Oh)

‘ ^

Q
o

&0 g

CD X
Q Qh
Q Q

•-H (U

— C/2

C3 n

S 3 s

>2 S

g Jf

3 3

u 3

ffi

CD
Q
C

_ 3

>r Q

CO Q

3 3

o H

S to"

S Q

S -Q

.5 o

to s ^

Q CO

3 W)

'Q H
Q OJ

g 33

o to

+->

<D Q
Q

O Q

Qh 3

r ,, '£ o

W T 3 Qh

(U ^

X Q-j

• i— H

P— *

ffi

-3*

-(->

* hH

JlT

i-H

C/2

-Q

C/2

# Qh
H-H

C/5

0 s

• in

’o

C/5

<d"

^ .
3

*H.

C3 O

co"

ot
O

02 ©
Eh
CD

O
.Q
_Q

D. X>

£ £

Oh)

C/5

ffi

«o

*H.

*H»4

> C(Q

-Q 3

02
o
o

w
<D

c i

§ ffi

o ■

2 )

l ' S

CQ co

Cf .co

b &

2 )

*H.

H-H

-a

C/3

0
a
a

C/3

C/2

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

+->

S-H

CD
c o

G G

. CD

M— I

X _T

S-H X

<D x
X S
□j G
^ GO

OO ^

On X

i — h CD

s H +->

_ ‘ CD

G X
-M O
CD CD

CCS

C/3

r--

03 fO

. CO
X 03

2 o>
O w

+-*

s ~

O £
CD g
& 2
C/5

00 s to
oo

03 W

CCS

+->

S-H

+->

CCS

-2

00 s

+->

/ \

2 r|

o £

■g 3

CCS 00

R jS

O O

5 GO
CD S
O -
« £

8 g

C/3

O tO

CD X

x E

GO <

-t->

S-H

C/3

CCS

G o

h >n

g‘ GO

' S

2 g

C/3

O oo

a, x

oo G O
S co
or- dh °
oo CD 03
OO — -I X
CD

2 CO

I *

CCS

+-*

c
2
3

O 3

- GO

H, 2

CD ^7— j • "

Q ^ 3

CCS

O cd
g X
o H

G X
2 G

<D G

G 'G

G (D

'cd G „

CD 'G CD

ft C S

M 2 -8

- S ^

^ S

OO <D

03 ^

^ G

x — ' CD

• CD

+-> CD

CD >

§ M $

rS G X

a s ^

CCS

_03

-c->

„ s

G GO

iS 1-

CCS

-c->

3 3

S M

Cl,

C/3

CCS

C/3
-(-* Ed

CD s

G 2
CCS

C/3

CCS

-6 2°
CD
X)

W oo

jX ^

6 ^

-h ri
m

« o 1

2 03 o

OO ■ o
o go in
m

ot-

- GO

8 *!

I <3

Ol- O
r i m
o ri
r-- 03 <n
— ; — ; rn

GO GO GO

^ ^ ^

•n" oC oT

•of

m
o o

O 03 03

GO GO GO

^ ^ ^

CCS

c§

H-H

CCS

C/3 / — s

X) t-i
9> «

c3 (D
CD >
G O
X)
CCS

o ■£
PQ §

C/3

H-l

S-H

C/3

S-H

G-h / — \
G <D
CD 00

rG g

C/3

CD c/3
c/3 G

G Gh
o

S-H

X
CD
G
O

m >H,

■' (D

> S

CD x

Q U

1 > <

T3 2

^ 2

2 s

3 S-H

£ £

_r °

"2 co

'gh 3

CD 2
G ^

& S

| g

§ S
E x
x

o
G
O

rs S

L) G

-d ^ s
X X
' G 2’ 3

sLs

"2 x

C3 <4— *

CO Ch
CD 2

■2 S —
S .SP 3

Gh G

- h CD

h t3

. CD
Gh

+->

CD
_ O

H G

X S'

CO CD

, X
r X

C/5

C/5 ^

X X)

>> Q

3 ^

o a

5s.
CD
ft

G i

G P
CD G

| S

I ^

X CD

OO G
O
O
r)

hR
CD

CD
CD
X
co

& 2

O CD

Cj 3

«o

c* o

Oft 03

CO U.R

G Oft

X X
2 ’g

X CD

• H co

2 ^
« CD

+->

-3 1

2
CD
G
X
OX) CD
co

0 X)

E =*

X c^
CD . r,

1 f

3 C2

Gh Ci

rr\

X X
K3 G

2 G

G >
G CD

in"

03
CD

D_t

2 o

Gh ^
00

x 3

G X
G X

S-H

o ■

C/5

5s.

+-H

•Ss

hR ri

R, O

R 03
tO -h

6JD

C/5

•*->

5s.

5d 03
OO

S5
CD CD

bft

«o

■x*

R ^

R O

Oo 03

'Sb S

CD OX)
R G
hR x

CD S-H

CS (D
R co

*■***«-

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

+-*

O £8

a C

<dv

G Cq
-I-* Tf

o vo

cd jR.
r|

8 2
c 2

CD ^

s A

.3 m
o vo

Td
Td £
CD S
-I-* X
o s-
(D <D

o W

— G

Cd Cd

'£ 2

2 o

Tl-

t|-

in co

o tI - co
CN O <N
oo cdv m
— ; r-

GO GO GO

2 2 2

l N

T|-

2*s

y—i

H— *

Tl-

ChO

"cd

Td"

ffi

vo"

# g

C/5

So "g

£ I

w £

£5

Td
CD
-(->
o

c/3

-t->

r-~

vo"

r-~

2 "o

G CD
CX

C/5

22 r !
2 2
< 2

< £

G OV

x g

>n .2
•n o
On ^

O °

*-H 2"N

to ’ — 1

Vh OV

0 vD
[X •
^ GO

ri ^

% c\

OV _

w OV

1 “

a £

>3 CD
Cd CD

S.f

-(->

>3 GO

‘C

.fa

8 w

-2

t|-

<dv

co"

<dv

8 2
2 v>

CD "
Oh

C/0

r-~

<N U
O (N
Tf O
in
ci m

2 2

o' o'
in

o ri

ov m

hh ri
2 2
2 2

O <dn
cd ’— ;

"2 2

CD 2

/ — s cd
oo •£
oo rx

2 *5

w ^3
_• G
cd G
tJ)
G

G ‘G
cd cd
M
o

c/o
t J)

’G ^

8 G
G o>

2 la

i-H

C/3

tJ)

G-h

0 ^

§ e

1 8

^2 G

2 u

^ c

to to 2

MG 3
G G CT

G Gh CD

g «* *

C/3

to ®

; cd g

G QJ
2 -
w

2 _

O go ;

C/3
G
G N

G £
G ' iJ
CG -G

Gh hG
G tcD

^cd

C-t— H
Gh

C/3

H — *

C/3

-G->

C/3

to'

C/3

~cd

,cd

c+t

2 )

C/3

+->

C/3

g O

w in’
m cdv

Oh ^
C/3 ^

R
o

G
Td
G
G

Td
G
to G
"1

J pq

-G

C/5

-X

*4-i

•—

"cd

C/f

C/3

a-)

J-H

a-)

•

Td"

>—

c/T

-M

"cd

f (D

H-H

C/3

a->

"cd

H—i

Tp?

O-

^ 2

1 I

H-H O
a> a

| 2

G H

G c/3

> cd

G ^

2' rn
2 ^
G cdv

C3 h

CO ^ ^
2 >3

§ §

S O

G Th

P-H .g

C/3

O-

A
(D

&
u

X3 £
g

cdv

G -M

tS «*

s-h Td
O <D
LLh +-*
^ G
>3 G
X) =P

Td §
G

2 2

8 2
g td

a £

c/3

cdv

C/5

H— »

C/3

2 id
G G
Oh O

C/3 .G

G
G
G
X)

cd ^

Td
Td
C/3 cd

C/3
G
G

HH ” X

>3 £ g^

2 o 2

-o cd av

S, >3 ov
M (D _H

y xj w

ffi O

o s

C/3

Td
G
cd

2
G
G
1 — ;

c§

G 'C
•G G
G 5/3
qd G
cd Td

£ £

C/3

+->

cd -G
h— * Td
G G

5 G

.2 Td

-o

DG ^

5 O

I i

c^ G

J G

"8 S'

-G oo
.2 ov

ffi o
O S

+~H

C/3

!/3

<dv

S-P

C/5

-(->

"C3

,fa

C/5

S—

-a

0
a

‘2

<dv

2 o

2 Ph

‘2

C/3

.=0

Proc. Linn. Soc. N.S.W., 136 , 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

—

C/5

CL)

CH-

T 3

CH-

G-

T 3

CH-

G-

* y

Ah Z

G- c3

CH- 8 )

o+ 2

^ G

A 'A

X CD

^ *-<
(—5 A
o (U
O H

^ CP
£ S

g -g

PQ £

CH-

-O

h-I

-O

h-I

& G

£ S>

C/5

5 xi

Gh o

> T 3

> G

i-H

C/5

-G A
o 33
G Gh
o *-h

r ) ^

U oO

i-H

"<d

C/5

o <D

3 -G

J s

CD CD

00 o

o 3 „G

to to
5G (D
G to

Q H

1 £

Gh _G
i-h G 3
CD o

% i

C/5

CD G
to HH
G 2
o

PC £

_h A
O ffl G
G „ G

O (D G

u G zn

C/5

CD G

do U

-G

M — 1

C/5

H->

C/5

"c3

e 2

i-H

C/5

H-*

"c3

'fa G

CD C3

C/5

^ 2
T3 -G
(D O

^ W

Q ^

C/5 C3

•2 a

O S-H

CD CD

G O

to w

G .2

to K

0 £
u *

n ■

<4)

<i)

’3

. 2 f

V-

C /5

0 /

Cl,

co
Ah
•

C /3

C 3

h§!

0 )

"Bh

C /5

C /3

-5

H-~

-5

s-T

R 3

C/5

0 /

-a

: G

•_

a no

'Q X?

Co On
Co 1—1

HS CD

t-i

55;

■2 A

T3
G
G
to
G
03

co 2
_<D W

’o J 5

0 >

-a

S-

-i

hC 3

G -

Co^ CD
to
CD

•R

<4)

<4>

-R

hS?

•et

T 3

&£

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

CL)

CH-

—

CL)

—

CL)

CH-

—

CO)

CH-

CL)

toD

to D

23 <d
5 a

CL <+h

i-H

C/3

-a 5

o PQ

5 c

E -5

^ EE
CD

CD Cd
3 to

•a cd
Eh a

CD H
an cd

fc g

E/2 a

i-H

C/3

> «2

cd a

Q S

„ <d

a; a

cd a

00 w

to <

3 1

a PQ

C/3

i-H

> |
<D a

Q S

i-H

■B

<D „

oo a

CD to CD

cd O

£ u

CD to

CO $D

=5
O

o
cd
O

o
o O

? £

g o

cd q

CD -i

a a<

, cd
Eh CD

<° W a

+-*

.S-H

"od

C/3

E £

"<D

+-*

g VH

a <D

"o "ed

o CD

cd a

td g

E 5

s -i
p s

a do

C/3

«o

-5 a

?s ^

•a

Ej
E3
Oh

C/3

<L>

s ^

(D O

Id Ei
cd a
a P

o
o

a cd
cd a
a a

£
a oo
eE eE
cd cd

C/3

S ^

^ S3

^ S,

a a

a co

g S

a a

o ee

o cd

>co

*§ a

§ 3

a o

s co

00
o
o
n

EE
cd
>co
cd

a
§

^ . .a

co >
=5 ^

.2 ^

$D cd

+->

O
(D -c - *

a od
o s

3 Jz

a ^

a cd
E o

e ^

H-H

•3

co
cd <D
cd a

O CD

23 a

o a

cd"

"E 5

tsi

2 ?

5 s.

Hs.

«o

* 5^

t-~

■a

«o

1 =
>E) cd

^ E

'S-

«o

Ee?

a
c5

hS:

«o

^H* __

hES CD
^ oo
^ D
CD E
Q CD

Oh PQ

5s.

«o

■SS

V-

«o

‘C

C/3

5s.

' — 1

.£f

_0D

2 2

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

CH-

T— H

i— i

t>c

C/5

l—! C/5

X X

0 CL,

S X

1 §
3 go

X -xo

« c

£ a

u 3

C/5

oUffl
^ ©
o

c/5 ©

O O-
Q t-i
H ©
w vi

C/5

5-h

CL)

G-h

CL)

x m

§3 clT

fC5 C/5

£ PC
3 ©

O X

S r8

X U

< u

C/5

(L>

H— >

^ .

C/5

G-

H— »

C/5

,Gh

Ghh

C/5

(L>

-X

C/5

J-h

C/5

‘G

C/5

'hi

■3

C/5

X ^

C/5 0>

0> C/5

C/5 ^

o 5
PC K

W X
X 9

-4—* 03

O O

a u

X C/5

S X

2 >

^ CD

O rN

■+-> XI
X o
(D X

Qh X
Gh Gh

(D X
V Oh

C/5

c/5

G-h

C/5

.©

'©

C/5

<D "O
> C

cs Cd
C/5 _h

a §

3 o

"? X
>3 CD
C/5 H-h

'■'H HH

° 8

?o X

x x

X bD

a x

c/5

C/5

3 s

.bp

(X)

’x

C/5

Ghh

‘G

C/5

+->

H— >

^^H

C/5

•|*»4

'©

G— hH

C/5

CXh

C/5

•|*»4

G-h

G-

C/5

G-

C/5

*l*«4

•1^4

Cl
05

£ 22

-3

■X

.Sh

^ a

* ^

'O G)

•-Z

►. ~

js tr

fj S2
£ -0
G)

5s.

_ Cj

s ^

C/5

’HH

o
o
o
5 <N

- § S

&a 30

a x

R X
'Sh ^

o £

Qh pq

-3

C/5

-X

-3

C/5

ST o

OG
Cj ’— 1

S S.

c/5

"©

C/5

•HS.

in
On
On
G) 1—1

G) IX
sS ^

’^»4 »— H

^5 --H

R =3

$» PP

X
X
X

X J M

sf &

^ .r" o

£ Q m

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

c/5

CL)

^ to

03
X ^
C (D

^ to

o
oo
x ^

C CD

—

'5b

y—i

y—H

y—t

i— i

_»

C/5

0-)

C/5

c/s'

H— »
c/5

C/3

£ —

§ ^

C/3

§ &

* C/3
T3 *5
CD

t3 <

= .£
O 73

° p
p p
CD

P o

•5 K

g T3
Ph p
C/3 c3

C/5

g Ph

p 22
x x

Ph 0

P CD
CD -P

& ‘i

. X
X^ c<3

X) i

TP ^

-HH O

o o\

_© h— H

22 w

• 2 C/3

5 S

1 <

6 p

00 CD
00

-P

p ^

P £

■§ E

O C/3

c/5

C/5

♦ r>

CP^

s — "

t/3

-C-H

C/3

X TP !_• '
<D CD

22 P "©

CD <D CD

’© 3 ^

PQ © ^

C/3

^ c3 <

S &
- t3

^ UP
to t'"~
P o
TP
CD

X g

b ©

co -P

" X
^ p

^3 w
"p &a

.3 S

■gf-S

O |

b a

5s.

A x

7 cd

. S-x

"O ^

• C/5

x g
On P
■ — i P
^ P

P &
X _P

^ X

3 x

P CD

X x

u ^

x S

X
CD

"G G

o —

co H
hS O

Q tin

ffl

_>3

H — *

’53

ffl

>x,

’b

C/3

+->

'sS

.£f

C/3

-a

C/3

o ■
o ■

ffl

O-

ffl

O-

'©

—

•HH

0 /

' 5 b

■HH

fee

L—

-HH

^Xi

C/3

1 h

.©

C/3

0 /

-n

-C->

-3

0 /

-a

-3

5s.

C/3

C/5

-a

-HH

C/3

^Xj

•0

-a

c- ■
o

■HH

§ e

S CM
to X

a On

•*-h.

§ P
hR X

5s.

1
P
Q X

c ^ U

-HH

-a

C/3

Jx i

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

c 3 ^

§ a

co
co

O
Gh

t-~

G-

CD OO

O °

« E

Oh,

«2 X

I <

6 c

00 CD

oo or!

rj\ CD
'-’V CD

w X

-I
“ §
G £

■§ 2

O GO

+-*

CL)

OX)

1—

f—

CL)

i-H

‘G

'+-*

t/T

too

O CD

S ^

w 5

i-H

X-»

c/5

+->

(L>

OX)

1—

i-H

-(->

’C

* ^

‘G

• G cd
<3 ^

2 03
2 CD
G -3

CJ CD
-G
03

s ~

=o
0)

Y- 3

® t3

^ O
X> *R

a X

CD O

3 Co

h : 'O

.So ov

J-O 3* w

•G ’P .2

X) r £> CD

2 ^ m

Gh 03

. s

T3
CD

won
•G ^ £
O £ 2

”3

X->

c/5

* ^ p^
G (D
^ Gh

*<*»4

(L>

ffi

C/5

cd ^

§3 ^

■8 §

X Cl

C/5

^ Uh

S W

*- 5

«o

<S)

«o

=o ?o

<S) X)

!■:«

S |
Q g

<S)

"cv -

^ s

'£* O

Js ^

C/5 I > ) / H—H

^ ^ On

l—

r<D

S-H

«o

I s

S .2

co" §

VD C
<3V G

G r3

G D

H-H

+->

•"HH

-s

co 00
G ■ — i

2 §

<o <N

R 2

I .S

I 2

| M
£> v

-§

s s

H-»

C/5

«o

Ctf

■a

-R

5 h.

C3V

‘S-

-2

■2 On
G 00

-s

4) - a

«S G^

s ^

o ^

-§ -a

« Q

SO Q

1^5

G-

<3V

-a

H-H

"Eh

s 2

«o

cSS

Proc. Linn. Soc. N.S.W., 136, 2014

INVERTEBRATE CAVE FAUNA OF JENOLAN

Proc. Linn. Soc. N.S.W., 136, 2014

S.M. EBERHARD, G.B. SMITH, M.M. GIB IAN, H.M. SMITH AND M.R. GRAY

Proc. Linn. Soc. N.S.W., 136, 2014

Jenolan Show Caves: Origin of Cave and Feature Names

Kath Bellamy 1 , Craig Barnes 2

’226 Rankin Street, Bathurst, NSW 2795, Australia, dbellamy@tpg.com.au
2 The School of Chemistry, FI 1, The University of Sydney, NSW 2006, Australia, cbarnes6@bigpond.com

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Bellamy, K. and Barnes, C. (2014). Jenolan show caves: origin of cave and feature names. Proceedings of
the Linnean Society of New South Wales 136 , 69-75.

The Jenolan Caves Historical and Preservation Society researchers and surveyors worked together to
place cave and feature names on maps being produced by the Jenolan Caves Survey Project. Their sources
for these names were guidebooks, newspaper articles, tourist publications, postcards, and photographs.
Valuable contributions also came from the oral history supplied by past and current guiding staff. From 1 838
to the present day, guides have striven to acquaint visitors with the “exotic” cave environment, resulting
in a tradition of giving features familiar names. To the informed, names of caves and formations can take
on a hieroglyphic character that can guide you through the history of the caves. Being aware of the feature
names can give a glimpse of the discoverers, prompt interest in the adventures of early visitors and even
recognise the work involved in making the caves accessible. The result is that over a thousand names have
been found that link historically and culturally the discoverers, management and visitors.

Manuscript received 26 August 2013, accepted for publication 19 February 2014.

KEYWORDS: cave lighting, cave names, Fish River Cave, Jenolan Cave.

In 2005, as part of the Jenolan Caves Survey
Project, the authors started to work on names for
the maps. The Jenolan Show Caves are made up of
“caves”, sections of a system that have been given
specific names to facilitate them as cave tours (Figure
1). The naming project immediately expanded as
some cave and feature names provided an historical
and cultural record of the Jenolan Show Caves.
The result is an important record of the tradition of
naming at Jenolan from the discovery of the caves to
the present day.

The early years

The first recorded descriptive names commenced
with the discovery of the arches in 1838 (Ralston
1989). Samuel Cook (1889) suggested that an arch
known as the Devils Coach House was so named for
reasons that had led to similar names for numerous
Devils Pinches and Peaks for surface features around
the world. Captain Cook had given the name Devils
Basin to a harbour because of its gloomy appearance,
being surrounded by savage rocks. For a brief period,
the Devils Coach House was renamed Easter Cave,
although the name never became popular.

The cave system has been known by various
names: McKeon’s Caves in 1856, Binda Caves in

1867, Fish River Caves in 1879, and finally on 19th
August 1884 the name Jenolan Caves was approved
(Havard 1933).

By the 1 860s names had been established for the
New Cave (Ralston 1989). Visitors began their tour
to this cave by hiking through the bush to Wallaby
Hole, entering the cave through the Sole of the Boot
to reach the Cathedral. They had to negotiate The
Slide by sitting on a bag and descending further into
the cave. In the Exhibition Cave they climbed over
rocks, lunched on Picnic Rock and drank water from
the Hidden River. In Lurline Cave those familiar with
William Wallace’s opera Lurline, first performed
in 1860, could see “...the coral bowers and cells to
which Rudolph was transported” (Cook 1889). In an
area of the Bone Cave called the Irish Comer there
was an interesting formation known as the Potato
Patch, and further along Bone Cave were Snowball
Cave and Crystal Fountain. Returning to Irish Corner,
visitors were astonished to find they had to ascend a
wire ladder to return to Cathedral and thence the cave
entrance. Although this route is not used today, many
of these names are still in use on the Lucas tours.

Some names became enshrined with the advent
of guidebooks; “English visitors see in this stalagmite
the features of Lord Salisbury” (Trickett 1905).

ORIGIN OF CAVE AND FEATURE NAMES AT JENOLAN

HI Jubilee

■ Imperial/Diamond
Elder
LJ Chifley
Nettle

Devils Coach House

I - ! Grand Arch

Lucas/Lurline
River/Pool of Cerberus
Baal

■ Orient
Ribbon

Figure 1. Jenolan Show Caves

Proc. Linn. Soc. N.S.W., 136, 2014

K. BELLAMY AND C. BARNES

Figure 2. The Old Curiosity Shop.

According to the 1924 Orient guidebook, visitors "...
one and all will recognise uncanny imitations...” and
decorations seem “...veiled in a film of suggestion
where more is meant than meets the eye and depends
to a certain extent upon the imagination” (Havard
1924). Figure 2 shows the Old Curiosity Shop where
such a process has resulted in 13 named features
amongst the mass of helictites. The names for the
features in this figure can be found in Figure 3.

At present the Orient (Figure 3) contains 134
named features, by far the most of any cave at
Jenolan. Many of the features have been renamed
over time, with some features like the Dome of St
Pauls renamed as many as 5 times (so far), to give
a total of 206 names for the Orient alone. There are
only 119 of the 206 names on Figure 3; it was not
possible to fit any more on!

Names and name changes

The reasons for names and name changes for
caves, parts of caves and features are multitude, and
the following paragraphs outline just a few examples.
The shapes that prompt a person to choose names are
usually explained by culture, history and, sometimes,
even profession. The chambers in the Orient (Figure
3) have names from that part of the world which is
now known as the subcontinent.

Upside Down Ice Cream Cone. A medical person was
probably responsible for describing the helictites in the
Dragon’s Throat in Baal as Diphtheria Symptoms.

The beautiful and small

There are many sparkling calcite crystal
decorations at Jenolan, such as stalactites, stalagmites,
flowstones and helictites, that have been named but
some of the most intriguing formations are obscure.
Old publications and photographs have enabled
identification of these treasures. Among one mass
of tangled helictites, named The Battlefield, is the
minute Leaping Stag. The Diminutive Horse Head
is one of the smallest examples of named features at
Jenolan (Figure 4).

The ambience of the environment

George Rawson (1882) wrote of a visit to Fish
River Caves that “...one is bought into a silent and
reverent attitude. . . ” hence it is no surprise that many
names of religious significance were used. There is
an Organ Loft and Pulpit in the Grand Archway, a
Sanctuary in Nettle, a Cathedral and Bishop in Lucas,
Twelve Apostles in Orient (Figure 5), with Imperial
and Chifley both having a Vestry. Biblical names
include Elijah’s Retreat, Tower of Babel and Lots
Wife

The imagination of guides and tourists

From the very beginning, cave guides and
tourists used names to describe formations, in part
to make the strange more familiar. It is a tradition
that has evolved and continues even to the present
as new cave is discovered. For example, renamed by
young visitors, the Unicorn’s Horn has become ET’s
Finger and The Minaret has become The Ice Cream
Cone. The cave divers have named a stalagmite as the

Historical events

Historical events have also played a part,
particularly in renaming features. The Terraces in
Exhibition Chamber became the Pink and White
Terraces in remembrance of those in New Zealand
destroyed by the 1886 eruption of Mount Tarawera
(Cook 1889). Mafeking was besieged during the Boer
War for 217 days, from October 1899 to May 1900.
The relief of Mafeking by the British from the Boer

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN OF CAVE AND FEATURE NAMES AT JENOLAN

& %
<

I £

** i C ,

* b
Si

I I

£ 1
-C Vi

,51

ft Si 5

i/I

1 "
cn

& £

'5 ^

i i! Q

C 6i g ♦«
^ Hi (U 1

-c %

r ^ ^ £

ca oi "p M

t- .£ pi ,*!

^ _£*.§ .3

<"Sl U. -J tx.

=C 4f

■-.■

Proc. Linn. Soc. N.S.W., 136, 2014

K. BELLAMY AND C. BARNES

yV;.

Figure 4. Diminutive Horse Head.

Figure 5. The Twelve Apostles.

coincided with the discovery of a high level passage
in the Exhibition Chamber, hence its name and the
names of some features in it (Figure 6).

Currently, there is a proposal to commemorate
the Queen Elizabeth II Jubilee with a named dome
and arch in Jubilee and Imperial respectively.

Figure 6. Map of Mafeking.

Honouring Australian dignitaries

In 1878 the New Cave was named Lucas after
John Lucas, M.L.A; “In consequence of the great
interest I displayed, and by the publication of my
paper, which first drew the attention of the public to
them, the Surveyor-General and other high officials
made an official visit, and named the largest cavern
The Lucas Cave” (Rawlinson 1976). One formation
was named Judge Windeyer’s Couch “.. .because it is
said that the learned judge sat on it when he visited
the caves” (Cook 1889). In 1952, the Left Branch
of Imperial was renamed Chifley Cave in honour of
J.B. (Ben) Chifley, who represented in the Federal
Parliament the region that included Jenolan. The
name change attracted some criticism; “The gesture,
however well intentioned, will not give much pleasure
to Mr Chifley’s admirers, for the sake of the memory
of a highly regarded man, I hope some more tactful
Chief Secretary changes the ludicrous name of Chifley
Cave back to what it was before” (Anon. 1952).

Cave incidents

Jeremiah Wilson, exploring Jubilee in 1893,
described the dreadful experience of having his candle
go out and believing he had no matches. Fortunately
he found some in his pocket, but he ensured the event
was not forgotten by naming the place where he was
at the time Wilson’s Despair. In Imperial, Ridleys
Short Cut was named after “...a visitor who stepped
back to allow a lady to pass and fell through (to a
cave below)...” (Leeder 1994). The guides describe
the incident as a “...rambling visitor who strayed from
the fold, put a foot in the wrong place, and descended
fifty five feet without the benefit of the rope. He
landed on a coil of netting and bounced off’ (Ralston
1989).

The influence of lighting

Different lighting can influence what can be

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN OF CAVE AND FEATURE NAMES AT JENOLAN

Figure 7. Queen Victoria.

recognised in the caves. Scenes lit by flickering
candles, and at times augmented by magnesium lamps,
delighted early visitors. However, the Stooping Lady
could “...be seen only by candle light, the magnesium
flare being too penetrating for this particular effect”
(Foster 1890). Harry Potter’s Scar was visible in the
Exhibition Chamber until the lighting system was
recently upgraded. The new lighting did however
result in a perfect representation of a Terracotta
Warrior appearing in shadow on the wall over the
River Styx in Lucas.

After ladies

In Lucas, Queen Victoria is unmistakable as she
looks out over the Royal Chamber (Figure 7).

Other ladies have figured in naming Jenolan features,
at times being substituted as their prominence
wanes. Queen Esther still has a chamber, Margarita
Cracknell, Selina Webb and Lucinda Wilson have
small caves, Katie Webb and Edie have their bowers,
while Josephine, Ethel and Minnie each still merit
grottos. Helena Hart cave was initially renamed in
favour of Lady Cecilia Carrington after her visit to
the caves, but this chamber is now called Madonna
Cave. Matildas Retreat, however, has become the
more mundane Marble Grotto. The provenance of
these names occasionally causes some dispute too. A
grotto attributed to Nellie Webb was challenged by
a visitor in May 1976 who stated the grotto “...was
named in honour of Nellie Carruthers...” (Mclver
1976) (Figure 8).

Conclusion

As part of the Jenolan Show Caves Survey,
the naming project has and will continue to assist
in providing accurate information on names and

locations, but it is an evolution. This is no better
reflected than in the words of a visitor who, in
September 1911, wrote on a Jenolan postcard “It is
wonderful how many shapes and images are suggested
to your imagination and I could have added hundreds
more...” (Anon. 1911). This discussion of the names
of Jenolan features and their sources has been
illustrated with only a few selected examples; it could
have been many more as the Excel spread sheets for
the Survey now contain more than a thousand names.
The Jenolan Show Caves can therefore be thought of
as the “Caves of a Thousand Names”.

ACKNOWLEDGEMENTS

The authors acknowledge the valuable contributions
from the Jenolan Caves Reserve Trust, Jenolan Caves
Historical and Preservation Society, the Jenolan Cave
Guides and the many caver helpers, Jenolan Caves Survey
Project for use of the maps, and A1 Warild for drafting
figures. Photos: C. Barnes (Figures 5, 7), J. Lim (Figure 2),
R. Whyte (Figure 4).

REFERENCES

Anonymous (1911). Jenolan postcard in Binoomea

No. 141. Jenolan Caves Historical and Preservation
Society

Anonymous (1952). ‘Candid Comment’, Sunday Herald
6 th April 1952, Sydney NSW.

Cook, S. (1889). Jenolan Caves: An Excursion in
Australian Wonderland, Eyre and Spottiswoode,
London.

Foster, J.J. (1890). ‘The Jenolan Caves’, (Charles Potter
Government Printer, Sydney - facsimile edition 2011,
Jenolan Caves Historical and Preservation Society).
Havard, W.L, (1924). ‘The Orient Cave’. (A. J. Kent
Government Printer, New South Wales - facsimile
edition 2011, Jenolan Caves Historical and
Preservation Society).

Havard, W.L. (1933). ‘The Romance of Jenolan Caves’.
(Artarmon Press, Sydney - limited facsimile edition,
NSW Department of Leisure, Sport and Tourism).
Leeder, P.M. (1994). Personal letter to Jenolan (archived
by Jenolan Caves Historical and Preservation
Society).

Mclver, F. (1976). Personal statement to Jenolan archived
by Jenolan Caves Historical and Preservation Society.
Ralston, B. (1989). ‘Jenolan: The Golden Ages of Caving’.

(Three Sisters Productions Pty. Ltd. Australia).
Rawlinson, N. (1976). ‘John Lucas: Conservationist or
Vandal’ (Occasional paper for 8 May 1976 meeting of
Jenolan Caves Historical and Preservation Society -
facsimile edition 2010, Jenolan Caves Historical and
Preservation Society).

Proc. Linn. Soc. N.S.W., 136, 2014

K. BELLAMY AND C. BARNES

Rawson, G.H. (1882?). ‘Guide to and
Description of the Fish River Caves’,
(original handwritten manuscript,
transcribed and illustrated 2013, Jenolan
Caves Historical and Preservation
Society).

Trickett, O. (1905). ‘The Jenolan Caves,

New South Wales’, 2 nd edition. (William
Applegate Gullick, Government Printer
Sydney).

Figure 8. Location of features named
for ladies.

Proc. Linn. Soc. N.S.W., 136, 2014

Understanding the Origin and Evolution of Jenolan Caves:

The Next Steps

R. Armstrong L. Osborne

Faculty of Education and Social Work, A35, The University of Sydney, NSW 2006

armstrong.osborne@sydney.edu.au

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Osborne, R.A.L. (2014). Understanding the origin and evolution of Jenolan Caves: the next steps.
Proceedings of the Linnean Society of New South Wales 136, 77-97 .

The dating of cave and surhcial sediments by Osborne et al. (2006) indicated that some sections of
Jenolan Caves, particularly the large chambers, formed in the Early Carboniferous before deposition of
sediments dated at 340 Ma. The dating also identified younger mass-flow sediments, dated at 303Ma and
secondary fine illite, dated at 258 Ma and 240 Ma indicating burial of the caves under the Sydney Basin.
These dates meant that a new chronology for cave development at Jenolan is required to supersede that of
Osborne (1996b). Construction of this chronology raises new questions: Did the paragenetic conduits form
before deposition or after stripping of the Sydney Basin? Caymanites (marine carbonate turbidite palaeokarst)
appear to be older than 340 Ma, but does this make palaeogeographic sense? The Early Carboniferous dates
give us a beginning for the history of the present caves at Jenolan, but much of the story is missing. Many
obvious features in the caves have not been studied. Present knowledge of the developmental history,
palaeokarst and sediment stratigraphy, morphology and mineralogy of tourist caves at Jenolan Caves is
insufficient to support sound conservation, management, development and interpretation. The next step in
understanding Jenolan Caves is a structured program of dating, geological, mineralogical and geomorphic
studies.

Manuscript received 17 July 2013, accepted for publication 11 December 2013
KEYWORDS: cave sediments, dating, Jenolan Caves, palaeokarst, speleogenesis

INTRODUCTION

Despite the popularity of Jenolan Caves, there
was very little study and very little was written
about the origin and evolution of the caves prior to
the publication of my synthesis (Osborne, 1999b).
Sussmilch and Stone (1915) speculated on the age
of the caves while Taylor (1923, 1958) attempted
to correlate cave development with that of the
Blue Mountains landscape using a fluvial model of
cave development. In the numerous editions of his
guidebooks Dunlop (1979) noted the role of solution,
cracks and the three streams passing through the
limestone in cave development. Beginning in 1983
I started a new study of Jenolan Caves, at first
concentrating on palaeokarst and the geological
record of cave development.

During the 1990s it became clear that while the
palaeokarst made sense, the morphology of the caves

themselves made little sense, particularly if they were
conventional stream caves as had been generally
accepted. After visits to Slovenia and Hungary in
1997, 1 realized that much of what we see at Jenolan
is quite unlike the text-book stream caves of Slovenia,
but the large dome-shaped chambers such as the
Temple of Baal have similarities with features seen
in the hydrothermal caves of Budapest. Looking at
the caves in a new light I saw both bottom up and
paragenetic features, which resulted in my first
attempt at putting the story of cave development at
Jenolan together (Osborne, 1999b).

Assumptions and definitions

In this paper I make certain assumptions about
the origin and evolution of Jenolan Caves and use
some terms in particular ways. Firstly, my basic
premise is that Jenolan is a multiphase / multi-process
cave system, which means that:

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

1 . Caves have formed several times in the
400 Ma history of the Limestone.

2. Some old caves are filled with lithified
sediment and are now intersected by
younger caves. I restrict the use of the
term palaeokarst to these sediments and
the features they fill.

3. Some caves contain very old sediment
contained within the same cave walls
that delimit the open cavities that it is
possible for humans to enter today. I call
these deposits relict sediments. I do not
use the term palaeokarst to apply either
to these sediments or to the cavities they
fill even though they may the hundreds
of millions of years old.

4. There are no simple answers to the
questions “How old are the caves?”
and “How did the caves form?” as
different sections of the accessible and
palaeokarst caves formed at different
times and by different processes.

Secondly, following Bella and Bosak (2012), I
have abandoned the use of the terms hypogene and
hydrothennal except where there is direct evidence
that hot water or water with a deep-sourced aggressive
agent is responsible for speleogenesis. In cases where
there is morphological evidence that a cave has been
excavated by rising water of unknown composition I
use the term per-ascensum.

METHODS

Morphology

Caves are underground landforms, so just like
surface landforms their gross morphology (seen by
visual observation, in plans and in long and cross-
sections) and their macro-morphology (seen in the
rock forms in the caves called speleogens) should
provide evidence for their mode of formation. In
the case of Jenolan the pattern of cave development
is strongly influenced by the shape and geological
structure of the limestone mass with passages north
of the Grand Archway following the general NNW-
SSE strike of bedding and cleavage and south of the
Grand Archway (“1” in Figure 1A) having a more
N-S orientation following a change in strike (Figure
1A).

In long-section (Figure 2) it can be seen that
while most of the cave development is horizontal,
there are specific zones of vertical cave development
spaced at apparently regular intervals along the

length of the cave. Osborne (1999a) recognised that
fluvial cave cross-sections in most textbooks showed
sections of caves in horizontally bedded limestone
(Figure 3 A) and that cave cross-sections in almost
vertically-dipping limestone like Jenolan would be
different (Figure 3B) and that paragenetic conduits in
vertically-dipping limestone would have a distinctive
cross-section (Figure 3C).

Three types of large solution cavities at
Jenolan can be identified on the basis of their gross
morphology; per-ascensum cupolas such as those
in the Mud Tunnels (“1” in Figure IB, Figure 4 A),
paragenetic conduits, such as that north of the Pool of
Reflections in River Cave ( “2” in Figure IB, Figure
4B) and fluvial streamways such as the Flitch of
Bacon ( “2” in Figure 1A, Figure 4C).

Morphostratigraphy

In caves like Jenolan where there have been
several distinct phases of cave development it is
possible to observe crosscutting relationships between
one cavity type and another. Recognising these
relationships can be a difficult and confusing exercise,
but should allow the relative ages of different groups
of cavities to be determined.

Sedimentology and Stratigraphy

Cave sediments can only be deposited after a
cave has formed and surface-derived sediments can
only enter a cave when an open pathway to the surface
exists. The age of the oldest sediment in a cave gives
the minimum age for the cave. The age of the bedrock
is the maximum age of any cave.

Figure 1 (NEXT PAGE)

A: - Plan silhouette of the Jenolan Show Caves
courtesy of Alan Warild, Jenolan Survey Project.
(1) Grand Archway; (2) Flitch of Bacon; (3) Tem-
ple of Baal; (4) Wilkinson Branch; (5) Katie’s Bow-
er, Chifley Cave; (6) Exhibition Chamber, Lucas
Cave; (7) Drain adjacent to Binoomea Cut; (8)
Ribbon Cave; (9) Jubilee Cave; (10) Pool of Cer-
berus Cave;

(11) Cathedral, Lucas Cave; (12) Bone Box,
Imperial Cave; (13) Imperial Streamway;

(14) Raft deposit in Imperial Cave (15) The
Mystery, Chifley Cave.

B: - Detail plan of River Cave area, omitting
Temple of Baal, Orient Cave and related cavities,
courtesy Alan Warild, Jenolan Survey Project. (1)
Mud Tunnels; (2) North of Pool of Reflections; (3)
Olympia Stairs; (4) Orient Stairs; (5) South of Ol-
ympia;

(6) T Junction; (7) Northern extension of Mons
Meg Loop; (8) The Ladder; (9) Mossy Rock.

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Proc. Linn. Soc. N.S.W., 136, 2014

sm m

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

g w 3

i i 3

o a
s B

-5 CTQ

o C/5
as 2

C/3

a 3
S3

pf- cz>

The grainsize and texture of cave sediments
and the sedimentary structures in them are good
indicators of the environment in a cave at their time
of deposition. Sand, small rounded pebbles, ripples
and imbricated cobbles are good indicators of fluvial
conditions. Mud, finely laminated and graded-bedded
layers and crystal raft deposits are indicative of a
lacustrine environment while mixtures of cobbles,
gravel and mud, without sand are indicative of mass
flow deposits.

Palaeokarst features and deposits are evidence
for the existence of caves in the past. Features with
bedding or other geopetal structures oriented to
the present horizontal must have formed after the
last folding event. Cave sediments and palaeokarst
deposits are difficult to date and can have very
complex stratigraphy (Osborne, 1984). This can
lead to the situation where even when an event is
dated, it can be of little help in understanding the
age relationship between major events assumed to be
younger or older.

Correlation

Ideally, it should be possible to correlate both
cave sediments and cave morphology with the known
geological and geomorphic history of the strata
and landscape in which a cave has developed. For
instance, incision events in the surface landscape
should correlate with incision and watertable lowering
in the caves. Erosion and deposition at the surface,
should, if there is a surface connection, correlate with
deposition in the caves. Major events in regional
geological history such as folding, granitic intrusion
and burial should also leave their mark in the caves.
In eastern Australia, however, correlation between
the caves and geological and geomorphic history
has proved to be neither simple nor uncontroversial
(Osborne, 2005, 2010). In the case of Jenolan,
the more we know, the more difficult some of the
correlation seems to become.

THE INITIAL SYNTHESIS

In my 1999 Presidential Address to the Linnean
Society of NSW I presented the elements of a
synthesis and a framework chronology for the origin
of Jenolan Caves. This recognized ten phases of cave
development; five phases represented by ancient
caves and palaeokarst deposits filling them, and
five phases identified by the morphostratigraphy of
and the sediments found in the presently open caves
themselves, Table 1 , below.

This chronology was largely based on
observations made in the southern show caves, which
proved to be more easily interpreted that those to the

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Figure 3, Passage cross-sections after Figure 17 of Osborne (1999a). (A) Textbook section of fluvial cave,
upper part of profile phreatic, developed below horizontal guiding joint or bed “G”; lower part vadose
canyon; (B) Cavity with similar origin to that in A, but developed along vertical guiding joint or bed “G”.
Note that vadose canyon is unchanged from “A”; (C) Cross-section of a paragenetic conduit developed in
vertically dipping limestone modelled after cross-section of passage at “2” in Figure IB.

Figure 4, The three main cavity morphotypes of Jenolan Caves. (A) Per-ascensum, ceiling cupolas in the
Mud Tunnels, River Cave, “1” in Figure IB; (B) Paragenetic, paragenetic conduit north of the Pool of
Reflections, River Cave, “2” in Figure IB, looking north. Note rising and falling notches in eastern wall;
(C) Fluvial, meandering vadose canyon, The Flitch of Bacon, Chifley Cave, “2” in Figure 1A. View look-
ing up to cave ceiling.

north of the Grand Archway. As no absolute dates had
been determined for either the clearly ancient material
or for the unconsolidated sediments in the caves, the
chronology was based entirely on stratigraphic and
morpho stratigraphic considerations and an attempt
to fit the cave chronology in with regional geological
and geomorphological history.

On these grounds I suggested that the palaeokarst
might extend back in age to the Early Carboniferous
Kanimblan Orogeny and that some cave filling,

such as the caymanites, might be Latest Carboniferous
in age, filling Carboniferous caves. Based on my
previous work (Osborne, 1995), I suggested that the
gravels on the surface at Jenolan and filling high-level
caves such as Dreamtime Cave were most likely to
be Permian in age. I recognized that the oldest phase
of development of the currently open caves was per-
ascensum development of the large cupolas such as
the Temple of Baal (“3” in Figure 1A). I thought
that this “phase 6” of cave development post-dated

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

Table 1. After Table 1 of Osborne (1999b)

A Framework Chronology for Jenolan Caves

Geological

Era/Period

Phase

Event/Process

Feature

Example

Present

Stability

Low Mg Calcite Speleothems

Orient Cave

Continued Weathering

Ribbon Cave

Mg Rich Minerals

Ribbon Cave

Quaternary

Meteoric Speleogenesis 5
Exhumation

Nick Point Sediment Cliffs

The Ladder, River Cave

Breakdown

Exhibition Chamber,
Lucas Cave

A number of

Cainozoic

Phases

Meteoric Speleogenesis 4
Paragenesis

Conduits

The Slide,
Lucas Cave

Loops

Mons Meg,
River Cave

? Tertiary

Meteoric Speleogenesis 3

Invasion Caves

Baal-River Passage

? Late
Cretaceous

Hydrothermal Speleogenesis 2
Hydrothermal Fills &
Alteration

Crystal-lined Cavities

Mud Tunnels,
River Cave

Dolomitic crystal

Pool of Cerberus
Cave

Altered Algal Mats

Ribbon Cave

Altered Palaeokarst

Olympia Steps,
Ribbon Cave

Non-Detrital Clay

River Lethe,
River Cave

? Late
Cretaceous

Hydrothermal Speleogenesis 2
Excavation

Cupolas

Persian Chamber,
Orient Cave

Halls

Jenolan Underground River

Tubes

Ribbon Cave

Permian

Cave Fill & Landscape Burial

Fluvial Sediments

Dreamtime Cave

Permian

Meteoric Speleogenesis 2

Large Caves

Dreamtime Cave

? Early Permian

Hydrothermal Speleogenesis 1

Crystal-lined Cavities

Lucas Cave Entrance

? Latest
Carboniferous

Marine Transgression and
filling

Caymanites

Olympia Steps, Ribbon
Cave

? Late

Carboniferous

Meteoric Speleogenesis 1

Phreatic Caves

Olympia Steps, Ribbon
Cave

deposition and partial removal of the Sydney Basin,
suggesting that it was likely to be Cretaceous in
age, resulting from hydrothermal activity related to
the opening of the Tasman Sea and the uplift of the
Eastern Highlands.

Just two years later, in March 2001, Horst
Zwingmann produced the first K-Ar clay dates from
Jenolan, and the whole world changed. Among the
first dates to emerge was the Devonian date (389 Ma)
for the sheared blue-grey clay from the Wilkinson
Branch (“4” in Figure 1A). This made sense as a
deformed palaeokarst deposit, correlated with the

volcaniclastics, which disconformably overlie the
limestone to the east, filling early caves.

The group of dates clustered around 340 Ma were,
however, a great surprise and puzzle. There were no
recorded Early Carboniferous strata within 1 80 km of
Jenolan Caves, the nearest being in the New England
Fold Belt (Figure 5), and it had never been suspected
that palaeokarst, cave sediments or strata exposed or
sitting on the surface in the Lachlan Fold Belt could
be of this age. The real surprise from the K-Ar dating
was that no Permian material other than overgrowth
crystals were found in the caves and that surface

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

150E

151E

Great Australian Basin

Sydney Basin

Talterang Group

Carboniferous Granite

New England Fold Belt

Lachlan Fold Belt

Figure 5, Regional geological setting showing location of Jenolan relative to Carboniferous strata.

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

deposits long thought to Permian, and represented
on geological maps as Permian, such as those in the
cutting on the Kanangra Wall Road at Mount Whiteley
turned out to be Early Carboniferous.

A CURRENT SYNTHESIS

If we use the K-Ar dating of Osborne et al. (2006),
recent observations in the caves and developments in
thinking about landscape development in and near the
Blue Mountains (e.g. van der Beek et al., 2001) to
modify the Osborne (1999b) chronology we end up
with Table 2 below.

Problems with the current synthesis

The lack of dating of events younger than the
filling of cupolas by mass-flow deposits, except for
the indication of burial under the Sydney Basin, makes
the present synthesis quite limited. While there is good
morphological evidence that cupola development and
filling was followed by a major phase of paragenetic
development there is no evidence yet as to whether
this event pre-dated or post-dated deposition of the
Sydney Basin, so I have represented this event twice
in Table 2, below.

Present knowledge does not allow correlation
between the cave record and the deposition of the
Sydney Basin, one of the major events in the regional
geological history. I, and many others, expected that
due to the proximity of the caves to the edge of the
Sydney Basin that basal Sydney Basin sediments
would be found in the caves. It is possible that we do
see the sediments in the form of the 303 Ma mass-
flow deposits in the Temple of Baal.

WHERE NEXT?

Geological problems outside the caves

Studies in caves are frequently impacted by
deficiencies in the basic knowledge of the geological
and geomorphic environment in which the caves are
located. There are several problems at Jenolan. While
the structure and composition of the limestone is
well known at a gross scale, more detailed structural,
stratigraphic and sedimentological studies would
help in understanding the factors influencing cave
development.

Dating some key features of the local geology
would also contribute to understanding the geological
background to cave development. It has been generally
assumed that the volcaniclastic rock overlying the
limestone is similar in age to the Devonian Bindook
Volcanic Complex, but this has never been confirmed

by dating the volcanics at Jenolan. Similarly, a range
of interpretations have been made about the age and
origin of the andesite located directly to the west of
the limestone near Caves House. These have ranged
from an Ordovician or Silurian submarine lava flow
to a Jurassic intrusion. Dating this rock would be of
great assistance.

To the southwest the sequence at Jenolan is
intruded by the Kanangra Granite and to the east by
the un-named granite into which Hellgate Gorge is
incised, both considered to be related to the Bathurst
Batholith. Pogson and Watkins (1998) stated that the
Kanangra Granite is likely to be middle Carboniferous
(325-330 Ma) in age based on general dating of the
Bathurst Batholith. They give the total age range
for emplacement of the Batholith as being between
340 and 312 Ma. The dates for the emplacement of
the Bathurst Granite overlap with those of the dated
clays given by Osborne et al. (2006) making it likely
the volcaniclastic source material for the clays came
from volcanism related to the emplacement of the
granite. As with the emplacement of the caymanites,
this presents a palaeogeographic problem. How could
the volcaniclastic debris enter the caves when at that
time they should have been covered by kilometres
thick of rock into which the granites intruded? Dating
of the Kanangra Granite and un-named granite may
help resolve this problem.

General problems in the caves

1. Underground cave/geologv relationships

Apart from some honours thesis work by
McClean (1983) and Allan (1986) and some small
scale localized work by David Colchester and me,
there has been practically no mapping of either the
bedrock and/or of the karst geology in the caves.
One factor preventing this was a lack of cave maps
of suitable quality and resolution onto which field
observations could be plotted. The recent completion
of the work of the Jenolan Survey Project means that
high resolution plans and sections are now available
for the whole of the show cave system.

Mapping the bedrock and karst geology of
the caves will make explicit relationships between
cave development bedrock lithology and geological
structures in the bedrock. It will also show the
distribution of palaeokarst features in the bedrock,
sediments filling the caves and the relationship
between speleothems, mineral deposits and bedrock
substrate. Unlike conventional cave maps, this type of
mapping wifi indicate were the cave wall is composed
of bedrock and where it is sediment, indicating the
outlines of sediment-filled cavities.

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Table 2. A Revised Jenolan Chronology

Relative

CAVE EVENT

CAVE

EXAMPLES

BEDROCK/SURFACE

Tertiary-Recent

Continuing fluvial action
and removal of old fills
Breakdown

Queens Canopy
Exhibition Chamber

Present surface streams

Early Tertiary

Active Streamways
Generation 6 Caves

Imperial Streamway
Lethe

Erosion 6 Extra Uplift of
Blue Mts. Inner valley?

Invasion meteoric caves
Generation 5 Caves

Baal-River Tunnel

Stripping of Sydney Basin

Erosion 5

Mid Cretaceous

100?

Uplift of E Highlands

Lacustrine & Calcite Raft
Deposits

Imperial

? Paragenesis
Generation 4 Caves

Mons Meg, Pool of
Reflections, Slide

Permian-Mid

Triassic

258-240

Secondary Elite Growth

Selina & Baal

Sydney Basin Cover

Latest Carb -
Triassic

Sydney Basin Deposition

L Carb-Permian

? Paragenesis
Generation 4 Caves

Mons Meg, Pool of
Reflections, Slide

Late

Carboniferous

303

Mass-flow sediments
with brown matrix

Baal, Orient, Imperial

Erosion 4

340-312

Post-Tectonic Granites

Mid

Carboniferous

320-327

Mass-Flow sediments
with yellow matrix

Erosion 3, Kanangra Rd &
Old School Diamictite

E Carboniferous

340

White & Yellow Clay

Baal, Orient, River

Volcanism

E Carboniferous

Per Ascensum 1
Generation 3 Caves

Baal, Orient,
Pool of Cerberus

Erosion 3

E Carboniferous

>340

Crystal vughs

River, Imperial

E Carboniferous

>340

Caymanites fill
Generation 2 Caves

River, Grand Arch
DCH

? Marine Transgression

Generation 2 Caves

? Crackle Breccias

E Carboniferous

Kanimblan Folding

Late Devonian

Unlikely to be found at Jenolan

Lambie Group

Unlikely to be found at Jenolan

Erosion 2

Unlikely to be found at Jenolan

Tabberabberan Folding

Late Early
Devonian

Volcanics overlying the
Jenolan Caves Limestone

Devonian

389

Blue clay palaeokarst

Wilkinson Branch

>389

Generation 1 Caves

Erosion 1

Latest Silurian

Jenolan Caves Limestone

While it is easy to see the benefits of such an
undertaking for cave management, interpretation
and science this project would require a considerable
amount of time and would require fieldwork by
experienced workers with eyes for carbonate geology,
structural geology, palaeokarst, cave sediments,
speleothem and cave minerals, hopefully working

in the field together, along with significant funds
allocated for lab work in petrology, structural geology,
x-ray mineralogy, sedimentology etc.

2. Age and origin of the crackle breccias

Crackle breccias consist of bedrock fragments in
a crystalline matrix. They are usually grain- supported

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

and often have the appearance of adjacent blocks
that have been pushed apart by the emplacement of
the matrix, and fit together like pieces of a jigsaw
puzzle.

There are two large exposures of crackle breccia
in the Jenolan Show Caves, both difficult to access
and sample. One forms the ceiling of Katie’s Bower
in the Chifley Cave (“5” in Figure 1A) while the
other is exposed in the cave wall and ceiling at the
bottom of the Slide in Lucas Cave at its junction with
Exhibition Chamber (“6” in Figure 1A). The Katie’s
Bower exposure (Figure 6A) shows evidence of
rotated blocks while the Lucas Cave exposure (Figure
6B) shows large angular blocks. Crackle breccias are
also found at Wombeyan Caves (Osborne, 2004) and
Bungonia Caves.

There are conflicting views about the origin of
this type of breccia. Polish economic geologists have
attributed the origin of these structures in dolomite to
solution-collapse following the removal of underlying
limestone (Sass-Gustkiewicz, 1974) while American
petroleum geologists (Loucks, 2007) have attributed
them to the collapse of cave systems due to burial
by an overwhelming mass of overburden. The latter
explanation seems most likely in eastern Australia.

While the Limestone was probably not covered
by a great thickness of Sydney Basin sediments, by
the end of the Devonian it was probably buried by a
significant thickness of mid-Devonian volcaniclastics
and siliceous late Devonian Lambie Group sediments.
While at present there is no direct evidence for the
age of these breccias, it seems likely that they are of
significant, possibly Devonian, age.

3, Age of the cavmanites

Unconformable caymanites (marine carbonate
turbidite palaeokarst, Jones, 1992) are exposed in
NSW in caves and in surface outcrop at Jenolan,
Bungonia and Borenore and in caves at Colong and
Wellington. While stratigraphic relationships suggest
they predate the Early Carboniferous clays at Jenolan,
they contain no datable macrofossils and attempts to
date them using microfossils have proved unsuccessful
as none were recovered. Palaeomagnetic dating has
been attempted with little success except to indicate
that they most likely predate the Sydney Basin.

Caymanite deposits are common at Jenolan in
the show caves, in the open arches, in the wild caves
and in surface exposure. One of the most important
exposures is at Olympia Steps in the Mud Tunnels
section of River Cave (“3” in Figure IB, Figure 6C).
Here an incomplete section more than 5 m thick is
exposed with a clearly defined unconformable upper
boundary, representing the palaeo-cave ceiling

(Figure 6D). The caymanite deposits include a range
of lithologies including beds of coarse crinoidal
grainstone (Figure 6E), graded-bedded sequences
(Figure 6F) and fine, cryptocrystalline mudstones.

The caymanites appear to represent an Early
Carboniferous marine transgression over parts of
the Lachlan Fold Belt, which is not recorded in the
conventional stratigraphic record. It is very difficult
to conceive an Early Carboniferous palaeogeography
that would allow marine water and sediment to
enter caves in the limestone at this time. The
palaeogeography of Late Carboniferous to Early
Permian times, however, is much more conducive to
such an event. So I (Osborne, 1999b) concluded that
the caymanites were likely to be Late Carboniferous
to Early Permian (Table 1). The problem is that
crosscutting relationships observed in the caves
by Osborne et al. (2006) and other examples seen
since all suggest that the caymanite is older than the
dated Early Carboniferous clays. Field evidence also
suggests that the caymanite is older that the crystal
filled vughs, which are also older than the dated
Early Carboniferous clays. Osborne (2007) discussed
the palaeogeographic problems arising from the
emplacement and survival of Early Carboniferous
sediments at Jenolan as part of the general problem
of explaining why ancient caves should survive at
all and suggested differential vertical movements of
fault blocks as a possible solution.

A new attempt at palaeomagnetic dating of
the Jenolan and other caymanites in New South
Wales and further studies of their stable isotope
geochemistry is planned and may help to resolve this
problem. Finding datable fossils or microfossils in
the caymanites would be the best outcome, but that
seems unlikely.

4, Effect of granite emplacement on the caves

While I have put a lot of thought into the
palaeogeographic implications of emplacement and
later un-roofing of the Carboniferous post-tectonic
granites for the survival of Early Carboniferous caves
at Jenolan, it was not until Dr Percival raised the issue
of “How did the granites affect the caves?” in his
presentation at the Jenolan Symposium that I thought
about whether I had seen any evidence that the caves
were affected by the emplacement of the granites.

Given that the boundary of the un-named granite
into which Hellgate Gorge is incised is 2 km east
from Jenolan Caves, and that the emplacement of
this granite was likely to have occurred between 325-
330 Ma, one might expect to see an impact on caves
older than 340 Ma and on the 340 Ma sediments
in these old caves. The emplacement of granites is

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

&«9

’•*#. -■}■ s&^wi ttfjnS

Figure 6, Crackle breccia and caymanite. (A) Crackle breccia in Katie’s Bower ceiling, note rotated
block in centre of image indicated by red arrow; (B) Crackle breccia exposed a western wall and ceil-
ing near junction of The Slide with Exhibition Chamber. Image courtesy Ted Matthews; (C) Olympia
Stairs caymanite exposure, looking south at “3” in Figure IB; (D) Upper boundary of caymanite deposit
representing ceiling of filled palaeocave in the Mud Tunnels near Orient Stairs (“4” in Figure IB) i = dip-
ping Jenolan Caves Limestone bedrock, ii = sub-horizontally dipping caymanite; (E) Exposure of coarse
crinoidal grainstone facies caymanite in Barrelong Cave, Lens cap 55mm; (F) Thin section of laminated
and graded-bedded caymanite from Olympia Stairs deposit.

10 mm

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

usually accompanied by significant heating of
the surrounding country rock, resulting in contact
metamorphism. In the case of the 340 Ma illite-
bearing clays one might expect this to result in the
growth of fine-grained spiky illite crystals during the
peak phase of granite emplacement between 325-330
Ma. We do find secondary spiky illite crystals on clays
from the Temple of Baal and on clays filling a crystal
vugh in Imperial Cave, but these give dates between
258-240 Ma, more likely to be related to burial under
the Sydney Basin than to the emplacement of the
granites.

Heating by batholiths often leads to hydrothennal
mineralization, and close to large bodies of limestone
could lead to hydrothermal cave formation and/or the
formation of crystal veins and vughs. Once again all
the available evidence suggests that the large per-
ascensum cupolas and the crystal vughs, both of
which could be hydrothermal in origin, are older than
the emplacement of the granite.

While the 12 km distance from the Kanangra
Granite might rule out any great impact from it, one
might expect an effect from the nearby un-named
granite in which Hellgate Gorge is incised. One
possible explanation for the apparent lack of impact
by granite emplacement on the caves could be that the
un-named granite is significantly older than 325-330
Ma. If the un-named granite was emplaced before 340
Ma, its emplacement could have been responsible for
both hypogene cave and crystal vugh development
without having any impact on the dated clays. This
idea could and should be tested by dating the un-
named granite.

A more radical possibility is that the rock mass
containing Jenolan Caves was not in its present
position relative to the granites at the time of their
emplacement, but was “shuffled” into its present
place by fault movements after the emplacement of
the granites but before the deposition of the Sydney
Basin. This is not completely impossible as there
is some evidence that the western boundary of the
limestone is faulted and House (1988) suggested
movement of the major north- south trending fault to
the east of the limestone post-dated emplacement of
the un-named granite. The relationship between the
caves and the granites remains a puzzle and work and
thought needs to be applied to solving this problem.

5, Age of gravels and mass-flow deposits

Dating by Osborne et al. (2006) gave two
different ages for the polymictic, matrix supported,
cobbly gravels at Jenolan Caves; approximately 320-
327 Ma for deposits on the Kanangra Walls Road
(Figure 7 A) and at the old school and 303 Ma for the

deposit that appears to have once filled much of the
Temple of Baal (Figure 7B).

Without the benefit of dating, Osborne (1995),
recognised that there were two distinct groups of
cemented gravels at Jenolan; polymictic gravels
with pyrite such as those in Dreamtime Cave (Figure
7C) and polymictic gravels without pyrite. It was
suggested that those with pyrite in their cement
were not Cainozoic in age and were most likely
latest Carboniferous to earliest Permian in age.
None of these gravels have yet been dated and their
relationship with either group of dated Carboniferous
mass-flow deposits at Jenolan or with other undated
gravels is not at all clear.

It is very likely that some gravel deposits result
from the re-working of older deposits. Some deposits
now on the surface may not be surficial deposits at
all, but deposits filling unroofed caves, such as the
gravel deposit on top of the Grand Archway (Figure
7D). A great deal of fieldwork in very steep country,
as well as in the caves, is required if any progress in
understanding the age and relationships of the gravels
is to be made.

6, Dolomite and ankerite

The Jenolan Caves Limestone is very pure and
in bulk contains very little magnesium. The caves,
however, contain significant isolated occurrences
of aragonite speleothems, often associated with
deposits of magnesium-bearing minerals such as
hydromagnesite and huntite and at one locality
dolomite is actively being deposited.

Ankerite veins protrude from the cave walls in
close proximity to aragonite deposits in Ribbon Cave
(“8” in Figure 1A), Jubilee Cave (“9” in Figure 1A)
and in the Mud Tunnels. Figure 8A shows protruding
ankerite veins at the southern end of Ribbon Cave
associate with a brown fill or alteration zone that
has yet to be sampled or investigated in detail. Also
growing from an apparently dolomitic substrate in
Ribbon Cave is a spectacular aragonite speleothem
mass called the Lyrebirds Nest (Figure 8B) with spiral
vermiform aragonite helictites tipped with growing
cauliflower-shaped masses of moist huntite with a
texture like cream cheese.

Some of the most impressive and extensive
aragonite speleothems occur in Pool of Cerberus Cave
(“10” in Figure 1A) associated with ferruginous mud
and soggy yellow weathered dolomitic limestone.
One section of the cave path has been cut through
some of the substrate to reveal yellow dolostone with
angular ferruginous fragments (Figure 8C). Some of
the aragonite speleothems in Pool of Cerberus Cave
and their rusty clay substrate are shown in Figure
8D.

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Figure 7, Gravel deposits. (A) Kanangra Road, tape marks unconformity at base of gravel deposit; (B)
Mass-flow deposit in western side of the Temple of Baal, Image courtesy Bojan Otonicar; (C) Cemented
gravel in Dreamtime Cave; (D) Gravel deposit, possible unroofed cave in saddle above the Grand Arch-
way.

The scattered deposits of aragonite and
magnesium minerals appear to be closely related
to ankerite veins and irregular dolomitic bodies in
the limestone. Some of the caymanite deposits are
dolomitized and it appears that a single bed towards
the top of the limestone sequence has been extensively
dolomitized. Weathered dolomitic/ankeritic net veins
can be observed in surface limestone outcrops. One
example is the veins exposed in the bank of the drain
running in front of the entrance to Binoomea Cut (“7”
in Figure 1A, Figure 8E).

Contact Cave, located high on the eastern side
of McKeown Creek valley, is named because it was
thought to have formed at the boundary between the
Limestone and the overlying Devonian volcanics. The
cave is close to, but not on the boundary and the rock
forming the eastern wall of the cave and much of the
ceiling is not composed of volcaniclastics but of rusty
yellow weathering dolomitic limestone. Complex
aragonite anthodites, with dolomite crystals forming
at their tips, grow from the weathering dolomite
substrate (Figure 8F).

Rowling (2004) described aragonite deposits in
several caves at Jenolan and suggested a relationship
with magnesium, strontium and sulfate ions, all of
which could be sourced from pyritic dolomite and
ankerite. Ross Pogson, David Colchester and I have
made some investigation of the ankerite and dolomite
veins and outcrops in the caves, but much more needs
to be done and funding is required for chemical and
isotopic analyses.

7, "Yellow stuff

Visitors and cave guides often inquire and
sometimes argue about the nature of striking yellow
coloured deposits partially Ailing or intersected by
the caves. These occur throughout the caves, but are
mostly noticed in the southern show caves. Now that
new maps are available it would be useful from both
a scientiAc and an interpretation point of view to map
and identify these deposits. Where these deposits
have been investigated the “y e U° w stuff’ turns out to
encompass a range of materials with a similar colour
and often a gooey texture. These include 340 Ma

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

Figure 8, Dolomite and ankerite. (A) Protruding orange ankerite veins and undetermined brown materi-
al on wall of Ribbon Cave (“8” in Figure 1A); (B) The Lyrebird, Ribbon Cave, a complex aragonite spe-
leothem mass with soft cauliflower-like deposits of huntite (indicated by red arrow) growing on the tips
of vermiform helictites. Black squares on scale 10mm; (C) Tan dolomitic mass with ferruginous clasts
intersected in excavated ceiling of Pool of Cerberus Cave (“10” in Figure 1A) adjacent to significant
deposit of aragonite speleothems; (D) Aragonite stalactites growing from ferruginous mud with curved
laminations (possibly weathered dolomite) in close proximity to “C”; (E) Dolomitic net veins in limestone
bedrock exposed in side of drain adjacent to entrance to Binoomea Cut (“7” in Figure 1A); (F) Aragonite
speleothems (anthodites) with dolomite crystals being actively deposited at their tips, Contact Cave.

clays, weathered ankerite veins, altered algal mats
and dolomitized diagenetic infill sediments with
bedrock fossils.

Figure 9 shows some examples of “yellow stuff’
from the southern show caves. Figure 9 A is one of
several crumbly sandy pendants that hang from the
ceiling of Pool of Cerberus Cave. This material is

clayey sand with no carbonate content and contains
small double-terminated quartz crystals, so it could
be Early Carboniferous volcaniclastic sediment.
Figure 9B is either a limestone boulder or a bedrock
projection from the cave wall exposed in the side of
a cutting in an old tourist path south of Olympia (“5”
in Figure IB). The rock has a thin coating of yellow

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Figure 9, Yellow stuff. (A) Ceiling pendant of siliceous “yellow stuff’ with old light fittings attached in
Pool of Cerberus Cave; (B) Undetermined yellow coating on exhumed boulder or cave wall in cutting
of old tourist path south of Olympia (“5” in Figure IB); (C) Dated Early Carboniferous volcaniclastic
sediment (orange) at T-junction in River Cave (“6” in Figure IB) Image courtesy Bojan Otonicar; (D)
Leisegang-banded ironstone with quartz grains, separated from bedrock by manganiferous reaction rim

on wall of the Cathedral, Lucas Cave (“11 in Figure 1A).

paste, which has yet to be analysed. Figure 9C shows
a bright orange remnant of dated Early Carboniferous
clay located at the “T” junction in River Cave (“6” in
Figure IB). Figure 9D shows a yellow ferruginous
remnant, consisting of a small number of quartz
grains in a ferruginous matrix, separated from the
bedrock by a layer (? reaction rim) of manganiferous

paste on the wall of the Cathedral, Lucas Cave (“11”
in Figure 1A). The origin and previous extent of this
deposit is unknown.

While in most cases the yellow colouring is
likely to be ferruginous, Ian Cooper pers. comm.
(2013) has reported observing native sulfur in both
River and Jubilee Caves, however this has yet to be

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

confirmed by sampling and analysis. Now that good
maps are available, a collaborative effort between
cave guides, marking localities of “ydl°w stuff’ on
maps and researchers sampling and characterising
the material is possible and could result in both better
interpretation and enhanced scientific understanding.

Southern show caves

The most important step in understanding the
history of the southern show caves is dating the
paragenetic sediments. These deposits are of two
types, sequences in wall niches and thick deposits
either filling passages or protected by flowstone caps.
The later type appear to be remnants of sediment
that probably once filled the whole length of these
conduits, exposed at the present erosion head.

Wall niche deposits are easily observed on the
niches in the walls of River Cave north of the Pool of
Reflections (“2” in Figure IB, Figure 10A). Sections
exposing sediments at erosion heads also occur in
River Cave. Sections are exposed at either end of the
Mons Meg paragenetic loop. An 8-metre section of
fine laminated mud (Figure 10B) fills what appears
to be the ancient northern route of River Cave before
its down-dip migration to the west (“7” in Figure IB)
while a section more than 6-metres high is exposed at
the Ladder at the southern end of the Mons Meg Loop
(“8” in Fig IB, Figure IOC). Another 8-metre section
is exposed at the northern end of the Mud Tunnels
near Mossy Rock (“9” in Figure IB, Figure 10D).

Northern show caves

Much of my work has focused on the southern
show caves as it is easier to study the cupolas and
observe morphostratigraphic relationships between
features produced by different phases of cave
development there. I had assumed, falsely as it
has turned out, that the northern show caves were
essentially stacked levels of fonner underground
streamways, filled with fluvial sediment, representing
a series of underground captures of McKeown Creek
(Osborne, 1999b).

What I have since realised about the northern
show caves is the difference in morphology between
the cavities along which the main tourist paths run
in Imperial Cave, Jubilee Cave and most of Chifley
Cave and the morphology of the cavity at river level
in the Imperial Stream way.

Near the main tourist paths the cave walls are
white and smooth. Scallops are rare and there is
little sign of sand (Figure 11 A). Cave morphology is
suggestive of excavation by paragenetic rather than
fluvial processes. Below, in the streamway, the walls
and projections from the ceiling appear to be made

of fresh limestone and are covered with many small
scallops, indicating fast-flowing water (Figure 11B).
In addition to the scallops, the rock surface is rough
due to the presence of small sharp pieces of insoluble
material projecting from the rock surface indicating
that the water in the stream is unable to dissolve small
pieces of chert and silicified fossils in the limestone.
There is clean sand with ripples in the streambed and
there are some overbank deposits of mud formed
during flood events. The active processes we see
today in the Imperial Streamway are clearly not the
key to the past as seen in the higher-level passages.

Recent casual observations have shown that
while there are relatively uncommon deposits of
fluvial sand and gravel, the principal sediment types
in the northern show caves are crystal rafts (Figure
12A), muds (Figure 12B) and poorly-sorted mass-
flow deposits (Figure 12C), indicative of lacustrine or
paragenetic conditions rather than fluvial.

While significant progress has been made in
unravelling the developmental history of the southern
show caves, there has been less progress in the north
and much remains to be done. There is a least one PhD
project in sorting out the sediments and morphology
in the northern show caves.

TAKING THE NEXT STEPS

Despite their ease of access the Jenolan Show
Caves are among the most complex and confusing
caves to study and understand. There are, however
very good reasons not just to persist with research at
Jenolan but to expand it. These include the scientific
significance of the caves, the significance of the caves
for interpretation and education, the significance of
the science for the conservation, management and
sustainable development of the caves, and their
natural heritage significance, which I believe could
be demonstrated to be at a level appropriate for
nomination to the World Heritage List.

Scientific significance

Jenolan Caves are among the world’s oldest
and most complex limestone caves containing
unconsolidated sediments dating back to the Early
Carboniferous and preserving records of past events
not found elsewhere. The caves are important
in illustrating the effects of multiple phases of
different cave fonning mechanisms, per-ascensum,
paragenetic, fluvial and breakdown being overprinted
within a small body of limestone.

The caves are also important for their great
diversity of mineral species and for the particular

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

S‘iC '

Figure 10,
Paragenetic Sedi-
ments.

(A) Mud deposits
on niches in eastern
wall of River Cave,
north of Pool of Re-
flections (“2” in Fig-
ure IB) wall approx.
6 m high;

(B) North extension
of Mons Meg section
8 m + (“7” in Figure
IB);

(D) Section at Mossy
Rock 8 m thick be-
low flowstone (“9” in
Figure IB)

Proc. Linn. Soc. N.S.W., 136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

Figure 11, Morphology of cave at tourist path level compared with that at stream level in Northern Show
Caves. (A) Imperial Cave tourist path, looking north, north of the Bone Box (“12” in Figure 1A). Note
relatively smooth walls and lack of scallops; (B) Looking down to the Imperial Streamway (“13” in Fig-
ure 1A) note scallops on ceiling at “i” and ripples in sand in streambed at “ii”.

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

Figure 12, Sediments in Northern Show Caves. (A)Calcite raft deposit in eastern wall of excavated tour-
ist path in Imperial Cave (“14” in Figure 1A). Pocket spirit level is 80 mm long; (B) Laminated mud
deposit near the mystery, Katie’s Bower, Chifley Cave (“15” in Figure 1A). Lens cap 55 mm; (C) Mass
flow deposit of cobbles and gravel in a mud matrix exposed in cutting of path to the Imperial Streamway
(“13” in Figure 1A).

Proc. Linn. Soc. N.S.W.

136, 2014

ORIGIN AND EVOLUTION OF JENOLAN CAVES: THE NEXT STEPS

expression of some forms of speleothem (see Pogson
et al. this volume).

Significance for interpretation and education

As Australia’s most visited show caves, with
some 240,000 cave visits annually, Jenolan Caves
are an important site for scientific and environmental
interpretation to the public, particularly for the
interpretation of Earth sciences. Of these visits,
11,700 annually are by primary and secondary
students, making it one of the State’s most important
school excursion venues.

Good interpretation requires a good story, derived
from rigorous theory, synthesis and a strong factual
base. For the caves at Jenolan we have a beginning
in the Early Carboniferous and an end in the present
cave environment; we know some of the events in
between, but not their sequence. Theory and synthesis
are now beginning to emerge, but as illustrated in the
case of “yellow stuff’ many obvious features of the
caves have not yet received serious scientific attention
and cannot be properly interpreted to the public.

Significance for conservation, management and
sustainable development

In order to properly conserve, manage and
develop a natural heritage site it is essential to know
what is there and if it is highly significant, rare,
fragile or vulnerable. Inventory studies did not exist
when Jenolan Caves were first developed for tourist
use in the late 19 th and early 20 th centuries, so our lack
of good data to inform conservation, management,
development and interpretation is partly historical,
but like most major show caves world-wide there has
never been an inventory study of the show caves at
Jenolan. Without an inventory study, monitoring of
caves is deficient (Osborne, 2002) so an inventory
study should be undertaken before any major changes
in cave management occur.

The work of Osborne et al. (2006), and the
continuing research proposed here has a focus
on unconsolidated sediments and less attractive
mineral deposits: materials that often receive less
care and regard during cave maintenance and when
development is proposed. Remnant sediment masses,
such as those near the Pool of Reflections could
easily be destroyed by over zealous use of high-
pressure water cleaning, while the first dated Early
Carboniferous clay locality was formerly used as a
source of material to repair drain pipes.

These ancient materials have, however proved
to be essential for understanding the history of cave
development and are records of past events not
previously known to science. The present risk at

Jenolan as in most other show caves is that something
of great significance might be harmed or destroyed
simply because it is un-recognised and un-recorded.

World Heritage significance

While Jenolan Caves are within the Greater
Blue Mountains World Heritage Area (GBMWHA),
neither Jenolan Caves, nor any of the other landscape
and geological features of the GBMWHA were
among the reasons for listing. There are many cave
and karst areas now included on the World Heritage
List so adding more would present a challenge.
However, there has been interest over many years
in the possibility of including Jenolan as part of an
Eastern Australian Impounded Karsts nomination or
in making a case to have the values at Jenolan Caves
included in the existing GBMWHA listing.

World Heritage listing requires places to be of
“outstanding universal value” and for non-living
natural places a detailed comparison of significance
with places having similar values internationally is
required. It is difficult to find caves internationally
with which to compare Jenolan, but I think there are
some caves in central Europe with which this may
be possible. A detailed understanding, listing and
evaluation of the values, and an inventory study would
be required. Any action on World Heritage listing is a
considerable undertaking and successful nominations
internationally always require the mobilization of
government and academic scientific resources.

CONCLUSIONS

There are clear steps to be taken to further our
understanding of the origin and evolution of Jenolan
Caves. Taking these steps is not only of scientific
importance, but will greatly enhance the conservation,
management and interpretation of Australia’s most
significant tourist cave system and is also essential for
progress towards World Heritage listing of Jenolan
Caves. The next steps require an application of cave
science at a scale not previously seen in Australia.
Are we up to the challenge?

ACKNOWLEDGEMENTS

This paper is an expanded version of a paper presented
at The Science of Jenolan Caves Symposium held at Jenolan
Caves on May 23-24, 2013. For the author, 2013 marks
thirty years of research into the geology, geomorphology
and mineralogy of Jenolan Caves. The ideas and some
of the images presented here have emerged from this

Proc. Linn. Soc. N.S.W., 136, 2014

R.A.L. OSBORNE

extended period of looking, puzzling, looking again and
just sometimes seeing the light. Firstly I must acknowledge
David Branagan who wrote the magic piece of paper that
gained permission for my first research trip to Jenolan in
1983, supervised my PhD, and always thought the caves
were old, but as it turned out not old enough.

Many Jenolan people have assisted with fieldwork,
paperwork, accommodation, and shared their valuable local
knowledge and insights with me. I must particularly thank
Ernst Holland, Nigel Scanlan, Andy Lawrence, the late
John Callagan, Andrew Fletcher, Stephen Riley, Stephen
Meehan, Ted Mathews and Dan Cove in this regard. It is
impossible to undertake research in show caves without
the cooperation and support of the cave guides and I must
thank guides past and particularly guides present for their
welcome, assistance and cooperation.

Understanding of Jenolan Caves has been greatly enhanced
by collaboration with mineralogy colleagues from the
Australian Museum: Ross Pogson and David Colchester
and revolutionized by collaboration with dating colleagues
from the CSIRO: Horst Zwingmann and Phil Schmidt.

The compilation of this paper has been greatly assisted
by the supply of maps and sections by A1 Warild, Jenolan
Survey Project, and the capture of a missing image by Ted
Matthews. My family Penney and Michael have endured
and survived my research and with her great eye for detail
Penney has read and corrected the drafts of this paper.

REFERENCES

Allan, T.L. (1986). Geology of Jenolan Caves Reserve.

BSc Hons thesis, University of Sydney, Sydney.

Bella, P and Bosak, P. (2012). Speleogenesis along deep
regional faults by ascending waters: case studies from
Slovakia and Czech Republic. Acta carsologica 41 ,
169-192.

Dunlop, B.T. (1979). ‘Jenolan Caves, 11 th Edition’. (New
South Wales Department of Tourism: Sydney).

House, M. J. (1988). The geology of an area centred on
Bulls Creek, Northeast of Jenolan Caves N.S.W. BSc
Hons thesis. University of Sydney, Sydney.

Jones, B. (1992). Void-filling deposits in karst terrains of
isolated oceanic islands: a case study from Tertiary
carbonates of the Cayman Islands. Sedimentology 39,
857-876.

Loucks, R.G. (2007). A review of coalesced, collapsed-
paleocave systems and associated suprastratal
deformation. Acta carsologica 36, 121-132.

McClean, S.M. (1983). Geology and cave formation,
Jenolan Caves, N.S.W. B.App.Sc. thesis, N.S.W.
Institute of Technology, Sydney.

Osborne, R.A.L. (1984). Lateral facies changes,

unconformities, and stratigraphic reversals: Their
significance for cave sediment stratigraphy. Cave
Science: Transactions of the British Cave Research
Association 11 , 175-184.

Osborne, R.A.L. (1995). Evidence for two phases of Late
Palaeozoic karstification, cave development and
sediment filling in southeastern Australia. Cave and
Karst Science 22, 39-44.

Osborne, R.A.L. (1999a). The inception horizon

hypothesis in vertical to steeply-dipping limestone:
applications in New South Wales, Australia. Cave
and Karst Science 26, 5-12.

Osborne, R.A.L. (1999b). The origin of Jenolan Caves:
Elements of a new synthesis and framework
chronology. Proceedings of the Linnean Society of
New South Wales 121, 1-26.

Osborne, R.A.L. (2002). Significance and monitoring.

Acta carsologica 31, 21-33.

Osborne, R.A.L. (2004). Tales from Marble Halls: The
geology and geomorphology of Wombeyan Caves. In
‘Caves and Karst of Wombeyan’ (Ed R. Ellis) pp. 55-
71. (Sydney Speleological Society: Sydney).

Osborne, R.A.L. (2005). Dating ancient caves and related
palaeokarst. Acta carsologica 34, 51-72.

Osborne, R.A.L, (2007). The world’s oldest caves: - how
did they survive and what can they tell us? Acta
carsologica 36, 133-142.

Osborne, R.A.L. (2010). Rethinking eastern Australian
caves. In ‘Australian landscapes’. (Eds P. Bishop
and B. Pillans) pp 289-308. Geological Society of
London Special Publication 346.

Osborne, R.A.L., Zwingmann, H., Pogson, R. E. and
Colchester, D. M. (2006). Carboniferous Clay
Deposits from Jenolan Caves, New South Wales,
Australia. Australian Journal of Earth Sciences 53,
377-405.

Pogson, D. J. and Watkins, J. J. (1998). ‘Bathurst 1:250
000 Geological Sheet SI/55-8: Explanatory Notes’.
(Geological Survey of New South Wales: Sydney).

Rowling, J. (2004). Studies on aragonite and its

occurrence in caves, including New South Wales
caves. Journal and Proceedings of the Royal Society
of New South Wales 137, 123-149.

Sass-Gustkiewicz, M. (1974). Collapse breccias in the
ore-bearing dolomite of the Olkusz mine (Crakow-
Silesian ore district). Roczink Polskiego Towarzystwa
Geologicznego 44, 217-226.

Sussmilch, C.A. and Stone, W.G. (1915). Geology of the
Jenolan Caves district. Journal and Proceedings of
the Royal Society of New South Wales 49, 332-348.

Taylor, G. (1923). The Blue (Mountain) Plateau. In ‘Pan-
Pacific Science Congress, Australia 1923, Guide-
book to the Excursion to the Blue Mountains, Jenolan
Caves and Lithgow’ (Alfred James Kent, Government
Printer: Sydney)

Taylor, G. (1958). ‘Sydney side scenery and how it came
about’. (Angus and Robertson: Sydney).

van der Beek, P, Pulford, A. and Braun, J. (2001).
Cenozoic Landscape Development in the Blue
Mountains (SE Australia): Lithological and Tectonic
Controls on Rifted Margin Morphology. The Journal
of Geology 109, 35-56.

Proc. Linn. Soc. N.S.W., 136, 2014

Geology and Geomorphology of Jenolan Caves and the

Surrounding Region

David F. Branagan 1 , John Pickett 2 and Ian G. Percival 3

‘Honorary Associate, School of Geosciences, University of Sydney, NSW 2006 (dbranaga@usyd.edu.au);
2 Honorary Associate, Geological Survey of NSW, WB Clarke Geoscience Centre, Londonderry NSW 2753

(picketj@bigpond.net.au);

3 Geological Survey of NSW, WB Clarke Geoscience Centre, Londonderry NSW 2753

(ian.percival@trade.nsw.gov.au)

Published on 30 May 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Branagan, D.F., Pickett, J.W. and Percival, I.G. (2014). Geology and Geomorphology of Jenolan Caves
and the Surrounding Region. Proceedings of the Linnean Society of New South Wales 136, 99-130.

Detailed mapping by university students and staff since the 1980s has significantly elucidated
previously poorly known stratigraphic and structural relationships in the Jenolan Caves region. Apart from
andesite of ?Ordovician age, rocks west of the caves probably correlate with the lower Silurian Campbells
Group. That succession is faulted against the Silurian (mid Wenlockian) Jenolan Caves Limestone, in
which caves developed during several episodes from the late Palaeozoic. Immediately east of Jenolan
Caves, siliciclastic sedimentary and volcaniclastic rocks with interbedded silicic lavas constitute the newly
defined Inspiration Point Formation, correlated with the upper Silurian to Lower Devonian Mount Fairy
Group. Several prominent marker units are recognised, including limestone previously correlated with
the main Jenolan limestone belt. Extensive strike-slip and thrust faulting disrupts the sequence, but in
general the entire Silurian succession youngs to the east, so that beds apparently steeply-dipping westerly
are actually overturned. Further east. Upper Devonian Lambie Group siliciclastics unconformably overlie
the Inspiration Point Formation and both are overlain unconformably by lower Permian conglomeratic
facies. Carboniferous intrusions include the Hellgate Granite with associated felsite dykes. The regional
geomorphology probably evolved from late Carboniferous-early Permian time, with ‘steps’ in the deep
valleys indicating episodic periods of valley formation, possibly including Permian glaciation.

Manuscript received 16 October 2013, accepted for publication 23 April 2014.

KEYWORDS: Carboniferous, Devonian, geomorphology, Jenolan Caves, palaeontology, Permian,
Silurian, speleogenesis, stratigraphy

INTRODUCTION

Jenolan Caves, located 1 82 km west of Sydney
by road (Fig. 1), are Australia’s best known and
most spectacular limestone caves. Early geological
studies concentrated on the narrow belt of limestone
and mapping of the cave system it encloses, whereas
more recent scientific research has emphasized the

speleogenesis of the caves and their antiquity. In
comparison, the regional geology surrounding the
Jenolan karst area has been relatively neglected,
largely due to its rugged terrain and structural
complexity. Thus the geological context in which
Jenolan Caves formed, that is so crucial to an
understanding of how the cave system evolved, has
taken a long time to unravel, and indeed still requires

GEOLOGY OF JENOLAN CAVES REGION

149°59'E

150°00'E

150°01'E

150°02'E

150°03'E

150°04'E

Allan 1986
Stewart 1987

Hallett 1988
House 1988

Road
Fire trail

— - Track

Drainage

0 IKm

Fig. 1. The Jenolan Caves region, showing the main access road and natural drainage pattern; inset map
shows location within the state of New South Wales. Also plotted are the outlines of student thesis maps
reproduced as Fig. 5 (Allan 1986, in purple), Fig. 6 (Stewart 1987, in blue), Fig. 7 (Hallett 1988, in green)
and Fig. 8 (House 1988, in red).

100

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

further study. Geological mapping by students and
staff (mainly at the University of Sydney) over the
past thirty years, notably during the 1980s, has greatly
improved knowledge of rock types, their distribution
and relationships in the vicinity of Jenolan, but this
work has remained largely inaccessible in unpublished
theses and field compilations. The results presented
here are primarily based on detailed field mapping
(at a scale of 1:10,000) and accompanying reports
by Allan (1986), Stewart (1987), Hallett (1988) and
House (1988) (Figs 1, 5, 6, 7, 8), supplemented with
mapping over the same period by D.F. Branagan
and K.J. Mills (all of the University of Sydney),
with additional thesis mapping by Doughty (1994).
Unpublished studies by Stanley (1925), Chand
(1963), Pratt (1965), McClean (1983), and E. Holland
(former Jenolan Manager) have also been taken into
consideration. Other geological investigations of the
area remain unpublished, although in preparation of
this paper we have had the benefit of discussion with
various workers (particularly Ian Cooper) who have
mapped the Jenolan Caves Limestone and nearby
strata in considerable detail.

PREVIOUS WORK

The earliest geological observations of a scientific
nature on the Jenolan Caves area were made by staff
of the Geological Survey of New South Wales (Fig.
2), including Wilkinson (1884) and Young (1884)
(vide Havard 1933). Fossils in the limestone attracted
the attention of Government Palaeontologist Robert
Etheridge Jr (1892) who described a pentameride
brachiopod and first assigned a late Silurian age to
these rocks. Initial usage of the name ‘Jenolan Cave
Limestone’ can be attributed to Etheridge (1894) but
the terminology was not formalized for another 77
years.

T.W.E. David (1894) (Fig. 2) concurred with the
late Silurian age of the limestone, also assigning that
age to the strata to the east that he later described (David
1897a) as consisting of “several hundred feet of dark
indurated shales, greenish-grey argillites, reddish-
purple shale and coarse volcanic conglomerates with
large lumps of Favosites , Heliolites , etc.”. David
(1897b) further postulated that “the felsite dykes east
of the limestone had assimilated much lime in their
passage through the limestone”, and suggested that
the conglomerates exposed on the Jenolan road 6
miles (9.7 km) from the caves were Upper Devonian.
Extrapolating from his work on similar rocks at
Tamworth of Devonian age, David surmised that the
cherty radiolarian-bearing rocks cropping out west

of the limestone belt at Jenolan were younger than
the limestone, for which he indicated a westerly dip,
and that their cherty nature was the result of contact
metamorphism by intruding dykes. Consequently he
revised his opinion of the age of the “Cave Limestone”
to Early or Middle Devonian, younger than the
limestones at Yass. However, David and Pittman
(1899), in a further examination of the radiolarian-
bearing cherty sediments, expressed uncertainty as to
whether the limestone was Silurian or Devonian.

Curran (1899) (Fig. 2) discussed some aspects of
the Jenolan geology, placing the eastern succession
of sedimentary rocks in the Silurian, intruded by
diorites, quartz- and felspar- porphyries, and included
a photo of one of the cuttings on the road down to the
caves.

The matter rested there until Morrison (1912),
carrying out a reconnaissance trip to complete the
proposed Geological Map of New South Wales
(published 1914), placed the limestone in the Silurian,
together with the adjacent rocks including ‘slates,
radiolarian cherts, claystones ... and contemporary
lavas’, and assigned a post-Devonian age to the
intrusive porphyries and felsite dykes observed by
David. Morrison (Fig. 2) noted the unconformable
nature of the junction between the Devonian
sandstones and quartzites [Lambian rocks] and the
‘Upper Marine’ Permian beds, and the occasional
occurrence of the younger strata abutting the
Devonian rocks, but several hundred feet below the
Devonian outcrops.

Sussmilch and Stone (1915), following brief
statements by Sussmilch (1911, 1913), presented the
results of a study of the caves region undertaken over
a number of years. Their paper remained the standard
explanation of the geology until at least the 1960s.
The study by Sussmilch and Stone (Fig. 2), based on
the outcrops along the Mount Victoria road (almost to
Inspiration Point), and the first few bends (Two Mile
Hill) of the Tarana road, the Six-Foot track and the
Jenolan River, recognised the essential lithological
variations. They dismissed David’s contention that the
cherts occurring west of the limestone were formed
by contact metamorphism and determined that they
did not dip conformably with the limestone, but
were probably brought into contact by overthrusting.
Sussmilch and Stone (1915) suggested that the cherts
(‘Jenolan radiolarian cherts’) and associated dark
claystones were of Ordovician age. They recognised
the ‘Cave limestone’ and the geographically distinct
(but then vaguely located) east-dipping ‘eastern
limestone’, which appeared to be unfossiliferous,
believing that the separated limestones belonged to
a single ‘bed’ on opposite sides of a large anticline.

Proc. Linn. Soc. N.S.W., 136, 2014

101

GEOLOGY OF JENOLAN CAVES REGION

Fig. 2. A selection of geologists who have made significant contributions (discussed in the text) to the
investigation or mapping of the Jenolan Caves region, spanning more than a century from 1884 to 1988.
The final four photographs are those of students from the University of Sydney whose B.Sc. (Honours)
thesis maps were used in the compilation of this paper. Some of the historic photographs are sourced
from Johns (1976) and Middleton (1991); others are from the image library of the NSW Department of
Resources and Energy. Top row (L to R): C.S. Wilkinson (NSW Geological Surveyor-in-charge 1875-
1891), T.W.E. David (University of Sydney), E.F. Pittman (NSW Government Geologist 1891-1916), Rev.
J.M. Curran. Middle row, left image: officers of the NSW Department of Mines c.1893 (clockwise from
top left L.F. Harper*, R. Etheridge Jr (Palaeontologist), O. Trickett (Inspector of Caves), M. Morrison
(Assistant Geological Surveyor); right image: C.A. Siissmilch (seated) and W.G. Stone, both of the De-
partment of Geology, Mineralogy and Mining, Sydney Technical College; Bottom row (L to R): J.E.
Carne (NSW Government Geologist 1916-1920), G.A.V. Stanley (graduate of the University of Sydney
1925), T. Allan (B.Sc. Hons 1986, S.U.), W. Stewart (B.Sc. Hons 1987, S.U.), M. Hallett (B.Sc. Hons 1988,
S.U.), M. House (B.Sc. Hons 1988, S.U.). *note that L.F. Harper was engaged on geological investigations
in areas other than Jenolan Caves.

102

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

Sussmilch and Stone grouped all the variably-
coloured, thin-bedded, highly-jointed rocks, east of
the ‘Cave limestone’, as Silurian slates underlying the
limestone. The unit identified by them as a rhyolite-
porphyry, cropping out close to the Grand Arch, was
an important marker for their structural interpretation
of an anticline, as it was located again west of the
eastern limestone. Other igneous bodies were
identified as intrusive. These included the andesite
occurring west of the ‘Cave limestone’, and quartz
porphyrites and felsites to the east.

The stratigraphic order set down by Sussmilch
and Stone (1915), in addition to their interpretation
of the geomorphic history, subsequently became
entrenched in the literature of Jenolan. Influenced by
Andrews’ (1911) concept of the Kosciusko Uplift,
said to have occurred at the end of the Pliocene, they
thought that the age of formation of the cave system
could only be less than 500,000 years.

The regional geological interpretation of
Sussmilch and Stone (1915) was accepted by Carne
and Jones (1919), who outlined more accurately the
outcrop of the limestone, showing it extending for
some distance both north and south of the tourist caves.
Of particular interest on their map is the marking of
two pods of limestone approximately 400 m westerly
of the almost continuous main belt at its northernmost
extent, close to McKeowns Creek, suggesting a
possible offsetting by faulting. However, mapping of
these bodies was clearly affected by the inadequate
base maps available to these earlier workers, as more
recent mapping shows that despite poor outcrop, the
limestone does in fact swing westerly away from
the creek, by flexure, and encloses these two pods.
Came and Jones also located the eastern limestone
more accurately than was shown on earlier maps (e.g.
Sussmilch and Stone 1915).

Sussmilch (1923) expanded a little on his
previous work, with a revised cross-section, and
provided a geological history beginning with a deep-
sea environment in which the radiolarian cherts were
deposited, shallowing to a warm sea in which the
eastern sediments were deposited, followed by clear,
shallow seas in which lime- secreting organisms built
up a mass of limestone (but not a reefal body). The
succession was thought to have been folded at the close
of the Devonian or during the early Carboniferous.

G.A.V. Stanley (Fig. 2) carried out considerable
mapping for an Honours thesis at the University of
Sydney in 1925, but this work was never published,
so the results were ignored for many years. He thought
the western succession was probably Devonian
with a gradational boundary against the limestone.
Though pointing out the considerable differences

between the main and eastern limestones, Stanley
still regarded them as stratigraphically equivalent,
interpreting the eastern body as closer to the sediment
source, while the main body he surmised to be of
reefal origin. He thus accepted the large anticlinal
structure suggested by Sussmilch and Stone (1915),
but believed it was complicated by cross faults and
strike -slip faulting. Stanley regarded all the igneous
bodies east of the main limestone as sills. Perhaps
Stanley’s major achievement was the preparation of
the first, surprisingly accurate, contour map of the
Jenolan region, using compass and tape, Abney level
and aneroid.

Geological interpretation of the Jenolan region
remained untouched for the next 40 years, until a
new round of university student studies took place in
the mid-1960s, with work by Chand (1963), Gulson
(1963) and Pratt (1965). However, stratigraphic
relationships and actual geological ages remained
uncertain and there was certainly some confusion
introduced by the incorrect assignment of bedding to
structures in the limestone and other units. Chand ’s
1963 work, extending a considerable distance east
from the limestone belt beyond Black Range, was the
most painstaking, including the collection of nearly
900 rock specimens from widespread documented
localities.

Branagan and Packham (1967) were the first to
recognise the overturning of the sequence east of
the limestone belt, and Packham (1969) revised the
stratigraphic relations, with the western units being
oldest, followed by the limestone and the younger
eastern beds.

Pickett (1969, 1970, 1981, 1982) provided much
of the modern palaeontological data available on the
limestone. His early reports identified macrofossils
submitted by C. Mitchell and T. Chalker who
remapped the limestone (Chalker 1971). Additional
fossils, predominantly corals, stromatoporoids and
algae, were identified for that paper by J. Byrnes.
The age of the Jenolan Caves Limestone was given
as Ludlovian (late Silurian). Conodonts diagnostic
of Silurian biostratigraphic zones, however, proved
elusive, despite 20 samples being processed from
throughout the extent of the western limestone belt
(Pickett 1981). Talent et al. (1975, 2003:198), citing
unpublished work by P.D. Molloy, mentioned the
presence of conodonts referable to the Ozarkodina
crispa Zone (of latest Ludlow age: Strusz 2007) from
the upper well-bedded part of the main limestone belt.
Unfortunately these specimens were not illustrated.
Strata underlying the Jenolan Caves Limestone were
assigned to ‘equivalents of the Campbells Formation’
(now Group) of late Silurian (Ludlow) age by Talent
et al. (1975: fig. 1, column 23).

Proc. Linn. Soc. N.S.W., 136, 2014

103

GEOLOGY OF JENOLAN CAVES REGION

Lishmund et al. (1986) presented a generalized
map of the limestone occurrences in the vicinity of
Jenolan and the immediately surrounding geology,
modified from mapping by Chand (1963), Gulson
(1963), Pratt (1965) and Chalker (1971). Lishmund
et al. (1986) also thought that the sedimentary rocks
west of (underlying) the main limestone belt might be
Silurian, based on lithological similarity to potential
regional equivalents, notably the Kildrummie
Formation.

Subsequent detailed mapping by Allan (1986),
Stewart (1987), Hallett (1988) and House (1988),
which has remained largely unpublished till now (Figs
5-8), forms the basis of our current understanding
of the geology of the Jenolan region, and is fully
discussed below. Osborne and Branagan (1985)
indicated a likely karstification age at least as old
as Permian for the development of the caves, and
subsequently (Osborne and Branagan 1988) included
a brief description of the Jenolan karst in an overall
review of karst in New South Wales. Detailed studies
of the Jenolan Caves Limestone, concentrating on its
karstification history, have been published by Osborne
(1991, 1993, 1994, 1995, 1999), Osborne etal. (2006)
and Cooper (1990, 1993). For a project directed to
developing tourism at Jenolan, Branagan (in Hunt
1 994) compiled a geological map based largely on the
detailed Honours thesis mapping undertaken between
1986 and 1988 mentioned above. Branagan et al.
(1996) summarized the results from this mapping
together with that of Doughty (1994).

Apart from the maps by Sussmilch and Stone
(1915), Carne and Jones (1919), Chalker (1971) and
Lishmund et al. (1986), other generalised maps of the
boundary of the limestone to have been published
include those of Trickett (1925), Shannon (1976),
and Kelly and Knight (1993), the latter which also
shows adjacent geology, based on unpublished thesis
mapping by Allan and Stewart. Osborne (1999)
in illustrating the limestone belt used mapping by
Shannon (in Welch 1976), but attributed it to Welch.

STRATIGRAPHY

Despite the rather formidable topography, some
excellent road and creek exposures can be measured
in the Jenolan area, providing the key to much of
the understanding of the stratigraphy presented in
this paper (Figs 3, 4). The road exposures were the
basis of the mapping by Sussmilch and Stone (1915),
although some sections of the roads have since been
relocated. Allan (1986) mapped the Inspiration Point
road section in great detail, providing the basis for

our revised interpretation of the rock succession
east of the Jenolan Caves Limestone. In Figures
5-8 depicting the detailed geology as mapped by
Allan (1986), Stewart (1987), Hallett (1988) and
House (1988), we retain the informal stratigraphic
nomenclature of their studies, but on the compilation
map (Fig. 4) the formal stratigraphic terminology
as described below is employed. It should be noted
that there are some differences apparent between the
compilation map (Fig. 4) and those of the student
theses (Figs 5, 6, 7 and 8). These differences are
due to additional field observations by Branagan
and K.J. Mills and consequent reinterpretation. The
main stratigraphic sections presented (Figs 5-8) are
Five Mile Hill to Jenolan Caves, Navies Creek, Bulls
Creek (with Pheasants Nest Creek), and the Jenolan-
Kanangra Road (Two Mile Hill section). These
sections reveal unequivocally that this stratigraphic
sequence is overturned (with but few exceptions)
— the succession younging to the east. Numerous
strike (and thrust) faults in the area separate the
rocks into distinct lithostratigraphic and structural
domains but because of the paucity of fossils so far
found, the relative age of these domains cannot be
stated with certainty. However, reinterpretation of the
scant palaeontological evidence provides the basis for
revised correlations of rock units west and east of the
main limestone belt, as well as reassessment of the
age of the Jenolan Caves Limestone.

A. Lower Palaeozoic rocks west of the main
Jenolan Caves Limestone belt

As indicated above, these rocks, with a few
exceptions, have generally been regarded as older
than the limestone and probably of Ordovician (or
alternatively Silurian) age. Pratt (1965) informally
referred to these beds as the ‘Oberon Hill Chert’,
including the andesitic volcanic unit exposed behind
Caves House (and in the Lower Car Park), which he
thought belonged within the ‘Jaunter Tuff’ of Shiels
(1959). Pratt noted closely-spaced concentric folds
within the chert sequence. Doughty (1994) followed
Pratt (1965) to some extent, informally naming the
succession of shales, siltstones, sandstones, cherts and
andesite, west of and underlying the limestone in the
vicinity of Jenolan, as the ‘Oberon Hill Formation’.
Doughty commented on the general lack of chert where
he examined the unit as the basis for the modification
of the name, and gave its minimum thickness as
between 1200 and 1500 m. The succession continues
north from the tourist area, beyond Dillons Creek (the
first main stream northerly from the Jenolan-Oberon
Hill road, draining from Oberon Hill and joining
McKeowns Creek opposite South Mammoth Bluff),

104

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAL

Ma.

Trough

Fig. 3. Palaeozoic stratigraphy and intrusion history of the Jenolan Caves region; cross-hatched areas
represent intervals of non-deposition and/or erosion (see text for discussion). Timescale from Gradstein
et al. (2012).

Proc. Linn. Soc. N.S.W., 136, 2014

105

GEOLOGY OF JENOLAN CAVES REGION

149°59'E

150°00'E

15G°01'E

150°02'E

150°03'E

150°04'E

- 33°44'S

1 Km

33 49 S

— 33°51'S

Fault

Road

REFERENCE

Recent

i Debris flow /
I alluvial fan

Permian

Conglomerate

Carboniferous

Microsyenite,
felsite dykes

Hellgate
Granite

Upper Devonian
Lambie Group

Quartzite

Undated intrusives

xxx

Hornblende

dolerite

Lower-Upper Silurian

Inspiration Point
Formation

Tuffaceous

sandstone

Spilite

Limestone

Rhyolitic

volcanics

Siltstone

Conglomerate

Interbedded

siltst./sandstone

Silicic & dacitic
flows

Quartz porphyry

Tuff & tuffaceous
sandstone

Lower Silurian

Jenolan Caves
Limestone

Lower- Upper Silurian
Campbells Group

Cherty

siltstone

Quartz

sandstone

? Ordovician

Andesite

-*■ Thrust fault s Fire trail

— Geological boundary ^ ' Track

— Anticline, Syncline Drainage

Chert / Mudstone
(age unknown)

Fig. 4. Geological map of the Jenolan Caves region, compiled by D.F. Branagan and K.J. Mills, based
on B.Sc. Honours thesis mapping especially as shown in Figs 5-8, and personal observations. Note that
there are minor inconsistencies between this map (which shows the formal stratigraphic nomenclature
adopted in this paper) and those of the students.

106

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

where Stewart (1987) mapped a sequence more than
500 m thick that he informally named the ‘Western
Jenolan Beds’.

1 . ?Ordovician andesite

The andesite (informally referred to as ‘Caves
House andesite’ on some maps), which has puzzled all
observers since the area was first examined, abuts the
Jenolan Caves Limestone over a short distance in the
vicinity of Caves House (Fig. 5). Chemical analysis
by Stone (in Sussmilch and Stone 1915) showed it
was originally of basaltic-andesitic composition. Two
rock types are present: a fine-grained augite-andesite,
and a porphyritic augite-andesite which occurs as
inclusions within the fine-grained rock. Chalker
(1971) suggested that the andesite represented an
intrusive body, although it has more generally been
interpreted as a flow, apparently conformable with the
limestone. However, it is probable that the andesite
unit has been brought into position by faulting along
McKeowns Fault and that its stratigraphic position is,
therefore, uncertain. Doughty (1994) noted that close
to Caves House, the Jenolan Caves Limestone contains
clasts of andesite, indicating an unconfonnable or
disconformable relationship with the andesite body.
Presence of an unconformity is supported by the
observation that in the eastern Lachlan Fold Belt,
andesitic rocks are characteristic of the Ordovician,
rather than the Silurian. Accordingly, the andesite is
most likely of Ordovician age, making it the oldest
rock unit exposed in the Jenolan region.

2, Campbells Group equivalents (Lower Silurian!

The ‘Western Jenolan Beds’ of Stewart (1987)
consist of two broad units, an older quartz-rich
sandstone unit, and a younger ‘cherty’ sequence (Fig.
6). The sandstone unit includes fine and medium-
grained sandstones, with very minor slates, and a thin
tuffaceous layer (possibly more than one). The unit is
dark to light grey with a distinctive blocky outcrop,
and occupies the ridge tops. In thin-section it is seen
to be composed mainly of rounded, strained grains of
quartz, with 5 to 10% of lithic fragments, and about
1% of mica fragments, and about the same volume
of matrix, composed of white mica, calcite, sphene,
chlorite and epidote. Iron oxide cement is present,
usually only about 1%, but in exceptional cases it
may make up about 20% of the rock, imparting a
dark colour to some hand specimens. The tuffaceous
layer is mainly composed of weathered felspar. This
sandstone unit continues west of the mapped area and
its thickness exceeds 450 m.

There is a distinct, but not sharply delineated,
lithological change to the overlying finer-grained

sequence. This sequence, about 500 m thick, is made
up of wide bands of thinly-bedded radiolarian-rich
black siltstones, interbedded with slates and minor
beds of quartz sandstone. The siltstone bands contain
tight slump folds, show graded bedding, small-
scale erosional features, and flame structures, which
indicate an easterly facing. In thin section the siltstone
consists mainly of a dark chlorite and quartz matrix,
with larger spheroids of microcrystalline quartz. These
are casts of radiolaria, often visible to the naked eye,
but they are generally poorly preserved and cannot
be readily identified. Occasional specimens display a
relict internal structure, and some bear short robust
spines.

Although evidence is slight in the immediate
vicinity of Jenolan, exposures to the north (in
McKeowns Valley) show that these ‘Western Jenolan
Beds’ have a faulted, and probably unconformable,
contact with the overlying limestone succession.
Sussmilch and Stone (1915) recognized an overthrust
fault, subsequently mapped by Stewart (1987) as
a high- angle reverse fault (the McKeowns Fault,
interpreted as a near-vertical thrust defined by a thin
layer of fractured rock) that separates this succession
from the Jenolan Caves Limestone. This fault is
noted also on the western end of the detailed section
measured by Stewart (1987) along Navies Creek (Fig.
6 ).

On the western side of McKeowns Valley there
is a 90 m wide zone of brecciation, consisting mainly
of cherty clasts (Fig. 6). Neither the displacement nor
the amount of strata missing can be determined, but
there appears to be no angular discordance between
the two units. However, in view of the apparent lack
of chert in the succession as it is mapped south to
Jenolan and beyond, it may be that this fault runs
slightly obliquely to the general strike of the beds
and cuts out the cherts. The width of the fault zone
certainly suggests that the effect of the fault could be
quite significant.

The ‘Western Jenolan Beds’ have previously
been assigned an Ordovician age by some authors
(e.g. Stewart 1987), although Pratt (1965) thought
they might range into the early Silurian. Packham
(1969) suggested that they could be correlated with
the Rockley Volcanics, cropping out to the west. Other
authors (Chalker 1971; Talent et al. 1975; Lishmund
et al. 1986) have regarded the rocks underlying the
main limestone belt to be of Silurian age.

Recent mapping by the Geological Survey of
NSW suggests that much of the Rockley Volcanic
Belt should now be regarded as Silurian, with reported
evidence of Ordovician ages (e.g. Fowler and Iwata
1995) from this tract to the west of Jenolan being

Proc. Linn. Soc. N.S.W., 136, 2014

107

GEOLOGY OF JENOLAN CAVES REGION

150°01'E 150°02'E

West Ridge East

REFERENCE

? Upper Silurian - ? Lower
Devonian

Eastern Beds

□ Spilite and
keratophyre

Quartz porphyry

Phyllites and
cleaved mudstone

Massive to impure
shaly limestone

Mudstone
Cleaved mudstone

Quartz sandstone

Upper Silurian - Lower
Devonian

Jenolan Beds

J Massive quartzo-
:• feldspathic sandstone

Interbedded

siltstone/sandstone

Conglomerate

Dacitic crystal tuff

Quartz porphyry

Cleaved mudstone

Upper Silurian

Jenolan Caves
Limestone

? Ordovician to Silurian

Black radiolarian-rich
mudstone

L. >

L. V

Andesitic lava

Fault

Geological boundary
Dip and strike

Dip and strike
inverted

Road

Fire trail
Track
Drainage
■ Huts

Fig. 5. Geological Map and cross section, modified from Allan (1986). Note that stratigraphic names utilized in the
Legend (except for Jenolan Caves Limestone) are informal.

108

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

33°46'S

33°47'S

33°48'S

A.S.L

1200m

1000m

800m

150°00'E

150°01'E

150°02'E

1 Km

REFERENCE

Permian

• 'A/.o .'*■**

AAA Conglomerate

Carboniferous

□ Micromonzonite
dyke

Upper Silurian to Lower
Devonian ?

Jenoian Beds

Dacitic volcanics

Turbidites

Sandstone

Quartz porphyry

Metabasalt

'Jenoian Rhyolite
Porphyry'

Cleaved slates

Upper Silurian
Jenoian Caves Limestone

Limestone outcrop
Limestone inferred

Ordovician ?

Fault breccia

Cherty siltstone

Quartz-rich
'• : •'••I sandstone

Fault

Geology boundary
Dip and strike

Road
Fire trail
Drainage

Fig. 6. Geological Map and cross sections, modified from Stewart (1987). Note that stratigraphic names
utilized in the Legend (except for Jenoian Caves Limestone) are informal.

reinterpreted as derived from allochthonous blocks
redeposited in the Silurian (C.D. Quinn, pers. comm.
2011). If so, this challenges the widely-held view
that the rocks west of the main limestone belt are
necessarily Ordovician in age, particularly in view of
the lack of fossil evidence.

Ordovician quartz-rich sandstones and cherts of
the Abercrombie Formation are extensively distributed
in the northern half of the Taralga 1 : 1 00,000 mapsheet

to the SW of the Jenoian area (Thomas and Pogson
2012). These homogeneous sandstones are described
as quartz arenites, with sublitharenites at the base
of the succession. Flame structures, flute marks and
load casts are present in some sandstone beds, and
the cherts frequently contain relict radiolaria (seen as
amorphous silica blebs). The presence of tuffaceous
layers in the ‘Western Jenoian Beds’ is atypical of the
Ordovician Abercrombie Formation. Therefore, it is

Proc. Linn. Soc. N.S.W., 136, 2014

109

GEOLOGY OF JENOLAN CAVES REGION

thought more likely that the ‘Western Jenolan Beds’
correlate with the lower Silurian to Lower Devonian
Campbells Group. This does not, however, explain
the age of the ‘Caves House andesite’ which remains
an enigma.

If, as surmised by Lishmund et al. (1986), the
strata west of the Jenolan Caves Limestone are
lithological equivalents of the Kildrummie Fonnation
(now included in the Campbells Group), this may
provide an age constraint on the overlying rocks
to the east. Conodont assemblages reported by de
Deckker (1976) led him to conclude a late Ludlovian
age (upper crispus to lower steinhornensis Zones) for
the upper Kildrummie Formation. Simpson (1995)
reinterpreted the specimens that de Deckker referred
to “Spathognathodns” crispus as Pa elements of
Kockelella ranuliformis, and thus suggested an age no
younger than basal siluricus Zone for the upper part of
the Formation. This age determination was influenced
by co-occurrence of other conodonts from the
Kildrummie Formation referred by de Deckker (1976)
to Diadelognathus primus and Distomodus curvatus.
As recognised by Simpson (1995), these clearly
represent elements of the apparatus of Coryssognathus
dubius, which ranges as high in the Yass succession
as the Hume Limestone, from which Link and Druce
(1972) recorded the zonal species Polygnathoides
siluricus. However, Kockelella ranuliformis first
appears locally in the amorphognathoides Zone
that spans the Llandovery-Wenlock boundary, and
typically occurs in the eponymous ranuliformis
conodont biozone of lower to mid-Sheinwoodian age
(early Wenlock). Its local upper limit was placed by
Bischoff (1986) within the K. amsdeni to K. variabilis
zones (late Sheinwoodian to mid-Homerian, or about
mid-Wenlockian). Thus it is likely that the age of the
upper Kildrummie Formation is no younger than mid
Wenlock. Correlation of the rocks immediately west
of the Jenolan Caves Limestone with the Kildrummie
Formation therefore implies that they are equivalent to
the lower Silurian portion of the Campbells Group.

B. Jenolan Caves Limestone

Prior to the present paper, the Jenolan Caves
Limestone (Chalker 1971) was the only formally
named stratigraphic formation in the karst conservation
area. This unit is dominantly a light to dark grey
bioclastic limestone, sometimes bedded, sometimes
massive, but it contains occasional mudstone lenses
and minor dolomite, and there is some evidence of
brecciation in places. The limestone outcrop extends
in a north-south linear belt for some 11 km. At the
northern end it is covered by younger rocks and
alluvium, but has become noticeably thinner, while
at the southern end it appears to have been cut off

by faulting. The succession shows some variations
in lithology as mapped by Osborne (1991) in the
Binoomea cut. Doughty (1994) suggested there are
four facies in the ‘southern’ limestone, including
(1) thin-bedded limestone and calcareous mudstone
(at the basal and top boundaries), essentially
lenticular and occasionally dolomitic); (2) massive
recrystallised limestone with thin mudstone partings,
forming the bulk of the Jenolan Caves Limestone; (3)
massive, discontinuous limestone composed of fossil
fragments; and (4) calcareous mudstone with minor
siltstone partings, which is intercalated with the other
three facies.

Doughty indicated a thickness of 350 m for the
limestone in the vicinity of Caves House, with a
considerably reduced section of 50 m in the south, and
noted that sudden reductions in thickness are due to
faulting which has usually removed the lower section
of the limestone. Just north of Dillons Creek, the
limestone is a maximum of 285 m thick and it thins to
75 m in the vicinity of Navies Creek. Here it passes
conformably upwards into limy shale about 5 m thick,
which in turn passes into cleaved slate, a few metres
thick, indicating continuous deposition, but with a
change in environment (and source). Alternatively,
cessation of massive carbonate production and
replacement by fine-grained mud-rich elastics may
have been caused by relatively rapid subsidence of
the carbonate platform below optimal water depths.
We regard these strata as the uppermost preserved
beds of the Jenolan Caves Limestone.

Cross-bedding and graded bedding can be found
in the limestone at one locality behind Caves House.
The cross-bedding has an eroded top, indicating an
easterly facing. This is supported by the overlying
graded bedding unit, which fines to the east.

Outcrops are variable, and particularly at the
northern end of the limestone belt it becomes difficult
to map the edge of the limestone accurately. In fact
although karst features, such as dolines, have been
used to map the western edge of the limestone, it
seems likely that the presence of large quantities of
water along the boundary may have caused dissolution
(or at least erosion and subsequent collapse) of the
adjacent “shales”, so the boundary of supposed
limestone may not be as accurate as one would wish.
The limestone dip is generally easterly, but close
to vertical, although it is often difficult to observe
or measure. The limestone shows considerable
topographic variation, which is in part the result of
facies variability, but also may be due to faulting, as
indicated by Shannon (1976).

In places the limestone is abundantly fossiliferous,
as seen at the entrance to the Binoomea Cut where
disarticulated pentameride brachiopod valves are

110

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

150°02’E

1 50°03'E

33°45’S-f

33°46’S -

33°47'S —

SW-

1100

1000 -

50 \

cc CO

4 55 1 o

{

Metres

(a.s.l)

1200

1100

1000

SW*-

1200 -

900-

V/H = 1

REFERENCE

Permian

Conglomerate

Carboniferous

Granite

Upper Devonian
Lambie Group

Quartzite and siltstone lens

Upper Silurian - Lower Devonian
Eastern Beds

— 7q

¥ Y-

Hornblende - dolerite
Siltstone

Tuff and tuffaceous sandstone

Spilite

Jenoian Beds

Rhyolite
Conglomerate
Silicic flow

Interbedded siltstone/
sandstone

Fault

Geology boundary

Dip and strike

Dip and strike inverted

Strike and dip of cleavage

Road

Fire trail

Drainage

NE-

Bulls

Wreckery

Creek

rvA

Fig. 7. Geological Map and cross section, modified from Hallett (1988). Note that stratigraphic names
utilized in the Legend are informal.

crowded in layers, perhaps representing storm
deposits. Elsewhere, corals (Favosites, Heliolites ,
Tryplasma, Phaulactis) and stromatoporoids
( Actinostroma , Clathrodictyon) occur sporadically,
previously regarded as indicative of a general
Ludlovian (late Silurian) age (Chalker 1971; Pickett
1981, 1982). Unfortunately, age-diagnostic conodonts
are rare in the limestone. The material identified by
Pickett (1981) includes Pa elements of Kockelella
ranuliformis (Walliser, 1964), illustrated in Fig. 9. As
discussed earlier, this species appears to have a much
longer range in Australia than elsewhere; nonetheless,
the youngest possible age is no younger than mid-

Homerian (mid-Wenlockian). This is the same
species which provides the best age control on the
Kildrummie Formation, so the conodont assemblages
do not pennit a differentiation in age between the two
units. Previously the most significant bio stratigraphic
information derived from unpublished work carried
out in the early 1970s by P.D. Molloy, subsequently
reported by Talent et al. (1975:64) and Talent et al.
(2003:198), that indicated the presence of conodont
assemblages of Ozarkodina crispa Zone age (latest
Ludlow) in the uppermost beds of the Jenoian
Caves Limestone. Regrettably, this material remains
unpublished. Endeavours to locate Molloy’s samples

Proc. Linn. Soc. N.S.W., 136, 2014

111

GEOLOGY OF JENOLAN CAVES REGION

150°02'E

150°03'E

150°04'E

REFERENCE

33'47'S

33°48'S -

V.E. = 1

Fault

Geology boundary
Dip and strike

Dip and strike
inverted
Strike and dip
of cleavage

Permian

y.o-\b'. ' Conglomerate

Carboniferous

■ — • — - Dykes

Felsite

Granite

Upper Devonian

Lambie Group

Quartzite

Upper Silurian -
Lower Devonian

Jenolan Beds

Silicic flow

Tuffaceous sandstone

Conglomerate

j Massive sandstone
Interbedded

siitstone/sandstone

Eastern Beds

Flornblende - dolerite

Interbedded

siitstone/sandstone

Tuff and tuffaceous
sandstone

Spilite

Silicic ? volcanic

Quartz sandstone

Quartz porphyry

Silt stone

Limestone

? Silurian

Voicanogenic

sandstone

Siliceous sandstone
and siltstone

? Ordovician

Chert

Road
Fire trail
Track
Drainage

Fig. 8. Geological Map and cross sections, modified from House (1988). Note that stratigraphic names utilized in the
Legend are informal.

proved fruitless, so they must be regarded as lost.
The age suggested by Pickett’s samples conflicts with
Molloy’s result, but since the latter can no longer be
checked, these should be disregarded. There appears
to be no basis at all for the assumption of a Pridoli
age for the Jenolan Caves Limestone, as claimed by
Scheibner and Basden (1998:478-479).

Halysitid corals have never been reported from
the Jenolan Caves Limestone, in marked contrast to
the Kildrummie Formation from which de Deckker
(1976:68) listed at least four species of halysitids.
Based on their absence, a comparison with the Yass
section thus implies an age at least equivalent to
that of the Hattons Corner Group (specifically the

112

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

Silverdale Formation) which lacks halysitids. Further
support for this correlation comes from faunas of the
1050 m thick Molong Limestone, from which Pickett
(2003) reported conodonts of the ploeckensis and
siluricus Zones, the boundary between these zones
lying 120 m stratigraphically above the last halysitids
in the section. These last halysitids are accompanied
by the rugosan Palaeophyllum oakdalense Strusz,
typical of the “Dripstone Fauna” of Strusz and
Munson (1997), to which they assigned an age range
of late Sheinwoodian to earliest Gorstian (i.e. mid-
Wenlockian to basal Ludlovian), approximately

ranuliformis to earliest crassa Zones. This accords
with the likely age for the Kildrummie Formation
deduced from conodonts (see preceding discussion),
and indicates that although the Jenolan Caves
Limestone is most probably younger, the difference
in age is slight.

In summary, taking the small conodont
assemblages as the most reliable indicators, but
considering the absence of halysitids, an age for
the Jenolan Caves Limestone near the top of the
Australian range of K. ranuliformis is most probable;
that is, mid-Homerian (mid-Wenlockian).

Fig. 9. Scanning Electron Microscope images of conodonts from the Jenolan Caves Limestone (A-J) and
limestone within the Inspiration Point Formation (K, L). A, C from GSNSW conodont sample C697; B,
D-J from GSNSW conodont sample C683 (for locations see Pickett 1981). Scale bars in all cases represent
100 microns. A-D, Kockelella ranuliformis (Walliser, 1964). A, Pa element in lateral view, MMMC4411;
B, Pa element in lateral view, MMMC4412; C, Pa element in aboral view, MMMC4413; D, Pa element
in oral view (note concentric growth lines around basal cavity), MMMC4414; E, Ozarkodina sp., Pb
element in lateral view, MMMC4415. F, G, J, Oulodus sp. F, Sb element in inner lateral view, MMMC4416;
G, Pb element in inner? lateral view, MMMC4417; J, M element in inner lateral view, MMMC4418. H-I,
Panderodus unicostatus. H, element in outer lateral view (unfurrowed side), MMMC4419; I, element in
outer lateral view (furrowed side), MMMC4420. K, specimen identified as the form-species Ozarkodina
ziegleri tenuiramea Walliser, 1964 by House (1988), MMMC4421; L, unknown coniform element in
lateral view, MMMC4422.

Proc. Linn. Soc. N.S.W., 136, 2014

113

GEOLOGY OF JENOLAN CAVES REGION

C. Lower Palaeozoic rocks east of the main
Jenolan Caves Limestone belt

The succession east of the Jenolan Caves
Limestone is complicated by faulting (Fig. 4). The
general stratigraphy (shales, lavas, and graded-
bedded sandstones) determined by Sussmilch and
Stone (1915) for rocks lying between the Jenolan
Caves Limestone and the Jenolan Fault - previously
referred to as the ‘Jenolan Beds’ by Allan (1986),
following Gulson (1963) and Chand (1963) - has
remained largely unchanged to the present, except
that the igneous units (felsic to intermediate types)
were variously interpreted as intrusives, while
other geologists regarded them as extrusives.
Equivalent rocks east of the generally north-south
trending Jenolan Fault were informally termed the
‘Eastern Beds’ by Allan (1986) (Fig. 5). These two
lithostratigraphic divisions were adopted by Hallett
(1988) (Fig. 7) and House (1988) (Fig. 8). Based on
mapping in the vicinity of Wombeyan Caves, 55 km
S of Jenolan, Simpson (1986) suggested correlation
of this succession with the Bindook Volcanic
Complex (now Bindook Group). The Bindook Group
is a variable association of volcanic, volcaniclastic,
clastic and carbonate rocks of Early Devonian age,
united by their silicic volcanic affinities, in particular
the presence of dacite. Outcrop of this association is
known to extend north as far as Yerranderie, 35 km
SE of Jenolan (Simpson et al. 1997). However, new
fossil finds reported here support a late Silurian age for
limestone interbedded with the steeply-dipping strata
overlying the Jenolan Caves Limestone. Accordingly,
we formally define a new stratigraphic unit, the
Inspiration Point Formation, that is characterized by
felsic volcanics and associated sedimentary rocks,
and correlate it with the lower to middle part of the
Mount Fairy Group, which is exposed in a NNE-
trending belt (the Goulburn Basin) on the eastern side
of the Goulburn 1:250,000 sheet (Thomas et al., in
Thomas and Pogson 2012).

Inspiration Point Formation (novT

Derivation of name', from Inspiration Point, the eastern
extremity of a prominent hairpin bend of the Jenolan
Caves Road below Mount Inspiration (Fig. 5).

Synonymy, the fonnation includes rocks informally
designated as the ‘Jenolan Beds’, the ‘Eastern Beds’,
the ‘Northern Beds’, and the ‘Eastern Limestone’.

Constituent units', no formal members are proposed,
but the formation includes several prominent marker
beds, including limestone, conglomerate, quartz
porphyry, and dacite.

Distribution', the formation extends from the eastern
margin of the Jenolan Caves Limestone to at least the
Black Range (eastern extent of the area mapped in
detail) (Fig. 4).

Geomorphic expression and outcrop', forms rugged
topography intersected by deep valleys; outcrop is
most accessible in creek beds and road cuttings.

Type area', due to the structural complexity, steep and
rugged topography, and heavily vegetated slopes, it is
not practicable to nominate a type section. However,
a type area can be designated north of Jenolan Caves,
bounded by the eastern margin of the main limestone
belt in McKeowns Valley and proceeding eastwards
across Binoomea Ridge to the main Jenolan Caves
access road (with good sections along this road
particularly between the Five Mile Hill in the Mount
Inspiration area and the Grand Arch at the Caves),
thence extending generally east from the main road
to Mount Warlock, and further north in the valley of
Bulls Creek.

Boundary relationships', the Inspiration Point
Formation is interpreted as conformably overlying
the Jenolan Caves Limestone, despite the sporadic
absence of a felsic volcanic unit at the base of the
formation (probably faulted out) that allows purple
and grey cleaved mudstone slightly higher in the
succession to abut directly the Jenolan Caves
Limestone east of the Grand Arch.

Thickness', a total thickness in excess of 3220 m is
estimated for the former ‘Jenolan Beds’, comprising
(from oldest to youngest) felsic volcanics to 30 m
thick; 350 m of purple and grey cleaved mudstone; a
prominent quartz porphyry with maximum thickness
about 150 m; an unspecified thickness of siliceous
mudstones interbedded with felspathic siltstones and
sandstones and containing a prominent conglomerate
bed 60 m thick; altered dacitic crystal tuff of 350 m
maximum thickness; a turbidite succession about 200
m thick in total; quartz-felspathic sandstone up to
145 m thick; turbidites about 85 m thick; a distinctive
crystal-rich tuffaceous sandstone about 200 m thick;
a further succession of turbidites about 1400 m thick;
and culminating in a series of massive volcanic rocks
with occasional conglomeratic lenses, more than 350
m in total.

Lithological variation: In the vicinity of Navies
Creek, grey slate at the top of the Jenolan Caves
Limestone is overlain by a band of felsic volcanics up
to 30 m thick forming the basal unit of the Inspiration
Point Formation. The volcanics disappear in the

114

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

south, about 400 m north of Dillons Creek. Outcrop
is patchy, but the unit probably sits directly on the
limestone south of Navies Creek. Lateral variation in
texture is rapid and common, and a lack of continuity
is not unexpected in such a unit, which consists of
fine ash, volcanic breccia with lapilli-sized fragments
and several flows of quartz porphyry and banded
dacite. The volcanic band is bounded on the east by
a shear zone, which may account partly for the lack
of continuity. Displacement on the fault, however,
seems to be restricted to the north, so that to the
south an apparently conformable contact probably
exists between the felsic volcanics and the overlying
unit, which consists of purple and grey cleaved slaty
mudstone 350 m thick. This is the unit which appears
to abut directly against the limestone just east of the
Grand Arch. When weathered it becomes quite red.
This unit contains numerous shear zones, but they
seem to occur only within possible bedding planes,
although bedding is rarely seen. In thin section this
mudstone consists of fine quartz grains in a matrix
of white mica, plagioclase and finer quartz. Doughty
(1994) indicated that a few thin beds of interbedded
volcanoclastic sandstones are present south of the
Grand Arch.

Occurring within the slaty mudstone is the
prominent quartz porphyry named ‘Jenolan Rhyolite
Porphyry’ by Sussmilch and Stone (1915), equivalent
to the ‘Binoomea Quartz Porphyry’ of Doughty
(1994), which is easily identified just to the east of
the Blue Pool. The porphyry occurs as two separate
bodies north from the tourist area, on about the same
stratigraphic horizon, but the more northerly body
approaches closely to the eastern boundary of the
mudstone and continues north with noticeable thinning
until about 500 m south of Navies Creek. South from
the tourist area, Doughty (1994) mapped the porphyry
as widespread around Green Ribbon Hill, possibly as
the result of folding (which apparently has affected the
limestone: Allan 1986), although Doughty suggested
that there are separate porphyry masses in this area.
The southern extent of the porphyry is obscured
under the Permian beds of Mount Whitely. In hand
specimen the porphyry is white with green patches
and contains large, fragmental phenocrysts of quartz
and felspar in an aphanitic groundmass, formerly of
fine glassy ash, now altered to chlorite, albite, calcite,
prehnite and sphene. The quartz crystals average
about 3 mm, and make up some 10-20% of the rock.
The felspar crystals (dominantly orthoclase), which
comprise about 15%, are altered, dull white and
usually smaller than the quartz phenocrysts. Biotite
is visible occasionally in hand-specimen. Evidence of
flow banding, pumice fragments (Fig. 10) and absence

of contact metamorphism indicates that the porphyry
is a primary pyroclastic rock (i.e. not reworked). Its
maximum thickness is about 150 m.

The purple and grey mudstone succession is
followed by slightly coarser siliceous mudstones with
interbedded fine felspathic siltstones and sandstones.
The siltstones consist of interlocking mats of fine
white mica and biotite with occasional very fine (<
0.1 mm) quartz grains. There is a dominance of grain
growth sub-parallel to bedding, probably reflecting
an original fissility. A prominent conglomerate bed
(60 m thick) within this package contains clasts of
limestone, spilite and mudstone ranging in size from
boulders to pebbles; cobbles and pebbles being
dominant. The matrix is relatively coarse sand,
composed of altered plagioclase (andesine) felspar,
calcite and spilite interspersed with finer grained
calcite, quartz, chlorite and white mica. South of
the Jenolan River the mudstone and siltstone units
appear to be conformable, but the conglomerate bed
swings northwesterly when crossing the Five Mile
Hill Road and runs directly into the purple-grey
mudstone unit. Allan (1986) attributed this swing to
faulting, and showed a fault of limited extent, striking
N-S, to explain the phenomenon, but the trend of the
nearby dacitic tuff (see below) shows a similar bend
and suggests that there might be a disconformable
boundary between the mudstones and the siltstones.
There is little or no evidence of an extensive
continuous fault along this boundary.

The sedimentary succession is interrupted by a
prominent dacitic unit, referred to by Allan (1986)
as an altered crystal tuff, but to the north interpreted
by Stewart (1987) as a flow. Its nature is concealed
by alteration. The unit crops out well as a distinctive
resistant band, weathering along joints and breaking
into large blocks. It varies in thickness from about
70 m in the vicinity of Navies Creek to 350 m two
kilometres to the south east. It is 120 m thick in the
Five Mile Hill road cutting, east of the Grand Arch,
and more than 300 m thick south of the Jenolan River.
These variations are probably largely stratigraphic,
although Stewart (1987) indicated that there is
evidence, in the form of brecciation of the dacite and
some sheared slates in a few places, that the eastern
boundary of the dacite may be faulted. The rock has
a characteristic pink-green groundmass, mottled with
dark green and light yellow-green patches, making
it readily identifiable in the field. The main primary
minerals are large grains of plagioclase, smaller grains
of quartz and finer quartz within the groundmass,
which also contains K-felspar. Chlorite patches
probably represent altered biotite. Granophyric and
micrographic textures suggest a flow rather than a

Proc. Linn. Soc. N.S.W., 136, 2014

115

GEOLOGY OF JENOLAN CAVES REGION

Fig. 10. Photomicrographs of thin sections of quartz porphyry (GSNSW T88825) from outcrop just east
of the Blue Pool at Jenolan Caves (identified as ‘ Jenolan Rhyolite Porphyry’ by Siissmilch and Stone
1915), showing the pyroclastic origin of this rock. A, with large pumice fragment in centre of field of
view; B, showing rounded volcanic quartz grains; C, with large plagioclase crystal on left side of field of
view, exhibiting twinning; D, with rounded volcanic rock fragment (dark grey speckled appearance) in
centre of field of view. Scale bar for A = 0.5 mm, for B, C, D = 1.0 mm.

pyroclastic origin. Alteration minerals, in addition to
chlorite, include epidote, albite, prehnite, pumpellyite,
calcite and white mica.

North of Navies Creek, and separated from the
dacite on the east by just a few metres of turbidites,
a thin layer of metabasalt crops out. It is a dark
green rock consisting of generally aligned, altered
plagioclase and clinopyroxene, partly replaced by
actinolite. Other metamorphic minerals — chlorite,
epidote, albite, prehnite, calcite and sphene — are
present.

The siltstone overlying the dacitic unit grades
up into medium-coarse grained quartz-felspathic
sandstones, with bed thicknesses varying from 5 to
50 m, often grading up from conglomeratic bases.
These pass easterly into well-bedded siltstones and
sandstones typical of turbidite successions, which are
about 200 m thick in total. Most of the sandstones have
volcanogenic sources and are composed of rounded
quartz grains (25%), altered plagioclase felspar (up
to 45%), lithic fragments (5-15%) and matrix. The

siltstones in this succession consist dominantly of
quartz, both as fragments and in the matrix.

The turbidites east of, and overlying, the dacite
are overlain in turn by quartz-felspathic sandstone
which is up to 145 m thick. It is coarse-grained, pink
and green, making it easily recognised. The southern
end of its outcrop, southwest of Mount Inspiration, is
cut off by the Jenolan Fault (see later). A succession
of turbidites about 85 m thick follows conformably.
Then follows a distinctive crystal-rich tuffaceous
sandstone about 200 m thick. This is a dark grey,
fine to coarse-grained inequigranular rock, consisting
of sub-angular grains of quartz, up to 5 mm across,
and euhedral altered white felspar (both plagioclases
and alkali types) up to 2 mm long, in a finer dark
cryptocrystalline groundmass, made up of quartz,
?albite and chlorite. There has possibly been some re-
working, so the unit is called a crystal-rich tuffaceous
sandstone rather than a crystal tuff. However, Stewart
(1987) suggested there is evidence that the unit grades
upwards from tuff into a quartz-felspar porphyry
flow.

116

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

A further succession of turbidites follows, about
1400 m thick, exposed along both the Jenolan Road
and in the various branches of Cookes Gully (Fig. 7).
It contains several mappable lenses of conglomerate.
The turbidites are interrupted, about 250 m above
the base, by what House (1988) and Hallett (1988)
referred to as a silicic flow. This rock is light grey to
pale yellow in hand specimen, aphanitic, marked by
black spots up to 2 mm across, and with numerous
fine pyrite grains. It was apparently originally a fine-
grained dacitic flow with phenocrysts of plagioclase,
mica and amphibole set in a fine glassy groundmass,
which devitrified to give fine quartz and albite. Later
low-grade regional metamorphism produced calcite
and chlorite.

The youngest unit of the ‘Jenolan Beds’ is
exposed along the Jenolan Road, where it has a
thickness of rather more than 350 m. The boundary
with the underlying turbidites is obscured by Permian
conglomerates, but it is probably conformable.
This uppermost unit is a series of massive, poorly-
layered volcanic rocks with occasional conglomeratic
lenses. The volcanic rocks range from siliceous
rhyolitic flows, sometimes with orbicular accretions,
overlain by a series of pink and green quartz-felspar
agglomerates.

A series of similar volcanic rocks, referred to by
Hallett (1988) as his ‘Northern Beds’, occurs east of
the Jenolan Fault along the Jenolan Road, and extends
along the Black Range Road. The ‘Northern Beds’
are separated from the ‘Eastern Beds’ by a northeast
trending fault, extending from the Jenolan Fault and
continuing at least 1.5 km to beyond the Black Range
Road (Fig. 7).

The stratigraphy of the area east of the Jenolan
Fault (previously referred to as the ‘Eastern Beds’) is
more complicated than to the west. This is the result
of faulting and the effects of contact metamorphism,
superimposed on regional metamorphism. In addition
the difficulties of access have made the interpretation
of the geology very challenging.

The oldest unit in this area is a small circular
exposure of buff white, intricately folded, laminated
chert, and associated fine-grained silicic sandstone,
found on a hillside 300 m north of Pheasants Nest
Creek. It may represent an allochthonous ‘window’
of material and appears similar to lithologies in the
Campbells Group west of Jenolan. Possibly associated
with the chert is a 1 5 m thick bed of silicic tuff which
crops out nearby.

Volcanogenic sandstone, overlain by siliceous
buff-grey fine sandstone fining upwards into siltstone
over several cycles, occurs along the valley of the
Jenolan River and the lower reaches of Bulls Creek,

and probably represents the next oldest strata in
the area. The sandstone fonns thick massive layers
with good outcrop, but exposure of the siltstone
is relatively poor. These beds are of undetermined
thickness and bedding is rarely readily identifiable,
but there is some evidence of younging to the east.
The sequence is cut off to the north by the shallow-
dipping Bulls Creek Thrust. This thrust has emplaced
a structurally overlying succession of limestone,
siltstone, spilite and tuffaceous sandstone which is
exposed in Pheasants Nest Creek, the upper reaches
of Bulls Creek and Beauty Gully (Fig. 8). All these
“upper” (?younger) beds have a distinct north-west
trend, which distinguishes them from the trend in the
‘Northern Beds’ (Hallett 1988), and in most of the
‘Jenolan Beds’, although the trend of the last named
does range from north-south to northwest-southeast.

A second belt of limestone, interrupted by
complicated folding and faulting, crops out about 2
km east of the main belt (Fig. 4). It extends below
Mount Inspiration on both sides of the Jenolan River,
and forms outcrops on Pheasants Nest Creek and
again on the north side of the Jenolan River south of
Beauty Gully. House (1988) mapped this ‘Eastern
Limestone’ in the form of a continuous body some
40 m thick (Fig. 8), rather than a series of isolated
pods as depicted by Carne and Jones (1919) and
Chalker (1971). However, on the south side of the
Jenolan River, upstream from the junction with
Pheasants Nest Creek, the limestone tends to occur
in the form of large lenses that grade vertically and
laterally into shaly sediments, as shown by Allan
(1986) (Fig. 5). The well-bedded shaly lower portion
comprises interbedded calcareous shales and massive
limestone layers ranging from 5 to 30 cm thick. Up
sequence the ratio of shale to limestone decreases
and it passes into a massive limestone, occasionally
developing small caves. Macrofossils are generally
not obvious in the limestone due to its pervasive
sheared appearance. House (1988) reported tabulate
corals, brachiopods, crinoid stems and gastropods
from one outcrop, and extracted from acid-insoluble
residues three conodont elements, one identified as
the form species Ozarkodina ziegleri tenuiramea
Walliser, 1964. However, reexamination of this
specimen (Fig. 9K) suggests that it is too incomplete,
with missing denticles, to be so precisely identified.
Recent fieldwork by Pickett and others, investigating
exposures of this limestone south of the Jenolan River
between Farm Creek and Pheasants Nest Creek,
led to recognition of the tabulate coral Propora, a
rugose coral identified as Pycnostylns (catalogued
in the Geological Survey of NSW Palaeontological
Collection as MMF45233) and large pentamerid

Proc. Linn. Soc. N.S.W., 136, 2014

117

GEOLOGY OF JENOLAN CAVES REGION

brachiopods similar to Conchidinm, all indicating an
age no younger than late Silurian.

The limestone passes upwards gradationally into
siltstone, about 50 m thick, with rare thin lenses of
quartz sandstone, the siltstone being succeeded by a
thick yellow-green to grey quartz porphyry up to 130
m thick, exposed in Pheasants Nest and Bulls Creeks,
and forming large pods to the east, below Warlock
Ridge. Both siltstone and limestone clasts have been
found within this unit. It appears to have an erosional
boundary with the underlying siltstone, but its upper
(northeastern) boundary is irregular and might be
faulted. A further thick siltstone with well-developed
cleavage, in reality a phyllitic succession, up to 360
m thick, ranging from purple- brown to grey and grey-
green, follows. It contains a prominent lens of silicic
?tuff. Towards the top of the siltstone succession there
are several lensoidal intrusions of hornblende dolerite
(see below). Thin layers of spilite are also present.
The siltstone unit is followed by a very thick spilite
(ranging from 500 m in the south to 900 m in the north),
which to the northwest is intercalated with tuffaceous
sandstones and siltstones. The thickness may possibly
be exaggerated by repetition through faulting. The
spilite is a dark green-grey rock of varying grainsize,
depending on the degree of recrystallisation, which
alternates between vesicular and massive types, with
evidence of auto-brecciation and pillows (up to 50 cm
across) towards the top of the unit.

North of, and ?faulted in places against the spilites,
is a thick sequence of massive grey coarse-grained
tuffaceous sandstones interbedded with laminated
siltstones. Some of the sandstone units are probably
scarcely reworked tuffs, containing occasional clasts
of siltstone and limestone. This sequence occurs in a
number of fining-up cycles, each 30 to 100 m thick.
Hallett (1988) indicated a total thickness in excess of
1200 m.

While thickness of each of the units varies greatly,
the stratigraphic succession remains consistent. The
succession is cut off by a north-north-west trending
fault, bringing it against Upper Devonian Lambie
Group rocks.

Age and correlation : Internal evidence of age of the
Inspiration Point Formation is meagre, being restricted
to the occurrence of Propora sp. and pentamerid
brachiopods. A constraint on the maximum age
of the Inspiration Point Formation is provided by
the underlying Jenolan Caves Limestone, of mid-
Wenlockian age. The Inspiration Point Formation
confonnably overlies the Jenolan Caves Limestone
and includes limestone that contains sparse fossils
no younger than late Silurian. Hence an age range of

latest Wenlockian or Ludlovian, possibly extending to
the Pridolian, is most likely for the Inspiration Point
Formation, correlating it with the lower to middle part
of the Mount Fairy Group described from the Goulburn
1 :250,000 map area SW of Jenolan. The Mount Fairy
Group in the Goulburn Basin ranges in age from mid-
Wenlock (early Silurian) to mid-Lochkovian (Early
Devonian). Thomas et al. (in Thomas and Pogson
2012) describe the lower to mid portion of the Mount
Fairy Group as comprising clastic sedimentary rocks
(including siltstone, mudstone and fine-grained
sandstone) and limestone lenses near the base,
interfingering with mainly felsic volcanics, consisting
of rhyolite, rhyodacite, dacite and andesite lavas and
volcaniclastic rocks. Graptolitic black shales, and in
other areas a succession of thick, regionally extensive,
fine- to very coarse-grained, quartzose to lithic-quartz
sandstone of turbiditic origin, interbedded with
siltstone and mudstone, overlies the lower portion of
the group. The upper portion of the Mount Fairy Group
overlying the turbidite sequence is characterised by a
thick succession of felsic to intennediate lavas and
volcaniclastic sedimentary rocks with minor basaltic
lavas. Thus there are considerable lithological
similarities with the Inspiration Point Formation.

D. Intrusive hornblende dolerite

The hornblende dolerite (mentioned above)
is a dark green, medium-grained holocrystalline
rock, the essential minerals consisting of dark green
hornblendes and white felspars with minor and
smaller green minerals which are probably epidotes.
The felspars (mainly albite) are up to 4 mm long, and
form an interlocking mass with coarse amphiboles
(to 5 mm), which makes up more than 50% of the
rock. Because the amphiboles are primary, the term
‘hornblende dolerite’ is preferred to other names,
such as amphibolite, which has the connotation of a
regionally metamorphosed rock. Using comparisons
with mafic rocks described by Joplin (1931, 1933,
1935, 1944) at Hartley, Macara (1964) suggested
that similar occurrences on the Kanangra Road
were associated with granite of Carboniferous age.
However, the occasional foliation which occurs in the
rocks here described, and even folding of individual
grains, indicates that these rocks are considerably
older, most probably pre-dating the Middle Devonian
Tabberabberan Orogeny (Fig. 3).

E. Lambie Group

Sedimentary strata assigned to the Upper
Devonian Lambie Group are medium to fine-grained
white-buff, well-bedded quartzites, quartz-rich
sandstones and siltstones (phyllites). Conglomerates,

118

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

which are typical of the basal Lambie Group in the
eastern Lachlan Fold Belt, are missing in the Jenolan
area, and have possibly been faulted out. Lambie
Group rocks have not been overturned but are folded
more broadly than the older units to the west, except
where the beds have been deformed adjacent to the
boundary fault, where they crop out in tight plunging
inclined folds. These rocks have been described by
Hallett (1988) and House (1988). While they would
not be seen by the casual visitor to Jenolan, they
occupy a significant place in the regional history of
erosion and karstification. Chand (1963) mapped
these beds extending well beyond Black Range, and
indicated the position of several fold axes of broad
folds. Hallett (1988) also mapped the more westerly
of these fold axes (Fig. 7), and noted the presence of
brachiopod fossils in the more phyllitic bands; these
indicate that the Lambie Group in the Jenolan region
was deposited in a marine environment.

F. Carboniferous Intrusive Rocks

The major intrusion in the area, here named the
Hellgate Granite, crops out on the Jenolan River about
3 km downstream from the Caves Reserve (Fig. 4),
and is equivalent in age (early to mid Carboniferous)
to the multiphase Bathurst Batholith (Fig. 3). The
edge of this intrusion was mapped by Chand (1963),
who regarded it as an offshoot of the Hartley Granite
(Joplin 1931, 1933, 1935), and by House (1988) (Fig.
8). Two phases can be recognized - a red granite and a
white marginal granite. The red granite making up the
main part of the body is a fine to coarse (up to 8 mm)
inequigranular, wholly crystalline rock. Pink-brown
coarse grains of quartz constitute more than 40% of
the rock, pink felspars consist of 35% plagioclase
and K-felspar 22%, with white mica making up
4%. The plagioclase is frequently altered. The white
granite is a medium-grained equigranular, wholly
crystalline rock with quartz (to 4 mm) comprising
35%, plagioclase (29%), K-felspar (25%), white mica
(up to 5 mm) (10%) and garnet (1%).

The contact with the country rock is sharp and
the granite roof, which is irregular, is often marked by
a 5-10 cm thick layer of coarse pegmatitic material.
A few smaller outcrops separate from the main body
occur upstream on the river.

In the south-eastern and topographically lower
portion of the mapped area, the Hellgate Granite has
caused noticeable contact metamorphism within the
‘Eastern Beds’ and the Lambie Group. The effects
of the contact metamorphism appear to be more
dependent on the depth of the granite below, than the
lateral distance from any granite exposure. A contact
aureole approximately 400 m wide has been mapped
(House 1988). In the inner 100 m an assemblage

characteristic of the hornblende hornfels facies
occurs. The outer 300-350 m of the aureole contains
an assemblage characteristic of the albite-epidote
hornfels facies.

A felsite dyke averaging about 10 m thick, first
mapped by Chand ( 1 963), runs northerly as an offshoot
from the granite, cropping out continuously for more
than 2 km to Warlock Creek, cutting obliquely across
the beds it intrudes. It is a pink-brown flesh-coloured,
equigranular fine-grained, wholly crystalline rock,
composed almost entirely of pink felspars (alkali
felspar 60%, plagioclase 30%, accessories 10%),
indicating a syenitic composition. Hallett (1988)
identified it as a syenite/monzonite where it crops
out at Warlock Creek. Several dykes identified as
micromonzonites were also mapped by Hallett
(Fig. 7). Another micromonzonite dyke, weathered
orange and dipping steeply SW, was mapped by
Stewart (1987) cutting NW across his ‘Jenolan Beds’
and the Jenolan Caves Limestone, north of Dillons
Creek (Fig. 6), and cut off by the McKeowns Fault,
thus antedating it [?post Carboniferous], It is 50 m
thick, mainly granular, but has some porphyritic
phases, and contains equal proportions of K-felspar
and plagioclase and about 5% of quartz and minor
groundmass.

G. Permian rocks

Conglomerates (with distinctive white quartz
pebbles) and sandstones, regarded as outliers of
the Shoalhaven Group (possibly equivalent to the
Megalong Conglomerate) of the Sydney Basin, crop
out sporadically. They occur mainly on ridge tops
on an old erosion surface, forming a plateau which
can be recognised extending far north to Mudgee
and beyond (Branagan and Packham 2000). In the
vicinity of Jenolan these sedimentary rocks occur
particularly along the Kanangra Road. However, there
are patches at various levels, sometimes lying directly
(unconformably) on the Jenolan Caves Limestone,
and very likely occurring also as cave fill in some
places (Osborne and Branagan 1985).

METAMORPHISM AND MINERALIZATION

Rocks of the Jenolan region are characterized by
low-grade regional metamorphism, mostly within the
greenschist facies range. In the pelites and tuffs of the
Inspiration Point Formation the regional pattern is
within the biotite zone in the upper greenschist facies.
Regional metamorphism has caused albitisation of
original basalts and andesites, producing spilites.
North of Jenolan, within the Inspiration Point
Formation, the sedimentary rocks occasionally fall

Proc. Linn. Soc. N.S.W., 136, 2014

119

GEOLOGY OF JENOLAN CAVES REGION

within the pumpellyite-prehnite facies, chlorite being
associated with green biotite, but a chlorite-epidote-
calcite-pumpellyite association is more common in
these strata. Actinolite occurs in a few instances in
doleritic rocks.

In the higher (topographically) country NW
towards the Jenolan F ault, contact metamorphic effects
diminish away from the granite intrusion at Hellgate
Gorge, but can still be recognized as an overprinting
on the earlier regional metamorphism. Occasional
retrograde metamorphism, marked by the occurrence
of laumontite, occurs within fractures and veins in
the spilites, and is probably attributable to circulation
of hydrothermal fluids. Minor mineralization (pyrite,
chalcopyrite and arsenopyrite), occurring pervasively
and in narrow veins, is possibly related to the
metamorphism.

Copper mineralisation (bornite, malachite and
azurite) associated with the spilites occurs in several
places in the Inspiration Point Formation. A little
bornite ore was extracted from a 20 m long adit early
in the 20th century (Carne 1908), and shallow pits
have been dug in malachite mineralisation in a 2 m
wide shear zone, where the malachite occurs in thin
veins throughout the rock and on cleavage surfaces.

STRUCTURE AND TECTONICS

The Silurian succession at Jenolan has been
structurally complicated (thus obscuring stratigraphic
relationships) by the effects of deformation during
three significant tectonic episodes: the earliest
Devonian Bowning Orogeny, the mid-Devonian
Tabberabberan Orogeny, and the early Carboniferous
Kanimblan Orogeny. The present attitude of the
Jenolan Caves Limestone and the Inspiration Point
Formation represents the combined effect of all three
of these orogenies. Upper Devonian Lambie Group
strata were affected only by the latter folding episode.
Permian strata are gently-dipping rocks, which post-
date the major folding and faulting. Folding of the
lower Palaeozoic succession is complex, with several
styles recognisable, restricted to different domains
that are separated by faults (most apparent, but some
interpreted). Within the newly-defined Inspiration
Point Formation (including Allan’s ‘Jenolan Beds’)
and the Jenolan Caves Limestone, Allan (1986)
mapped a series of large scale, open and fairly
symmetrical, near-recumbent folds (wave lengths of
the order of 400 m), the fold axes plunging northerly,
on which smaller-scale parasitic folds (wave length
of less than 40 m) are superimposed. To the east there
are large-scale anticlinal structures, gently plunging

north-easterly, on which are developed (at outcrop
scale) both asymmetrical kink folding and fairly tight
symmetrical folds. In some areas both cleavage and
bedding can be clearly seen to be folded. Allan ( 1 986),
Hallett (1988) and House (1988) deal in considerable
detail with the complexities of folding in the region.

Thrusts, or steeply-dipping reverse faults, dipping
both east and west, are probably extensive. North-
south striking vertical faults, probably in part strike-
slip, are also common. The major (and some minor)
faults mapped or interpreted are shown on Fig.4.
Several of the faults are of regional significance, in
particular the fault bordering (or close to) the Jenolan
Caves Limestone on the west (Stewart’s McKeowns
Fault), and the Jenolan Fault striking generally north-
south just west of Mount Inspiration.

The Jenolan Fault separates the two
lithostratigraphic and structural domains previously
informally termed the ‘Jenolan Beds’ and the
‘Eastern Beds’. Its outcrop pattern indicates that it is
consistently close to vertical. Although Allan (1986)
believed this was a high-angle thrust fault, House
(1988) presented evidence that it was more likely a
dextral strike-slip fault with some nonnal component
of displacement. The evidence is of two types:
shallowing and bending of cleavage in the ‘Eastern
Beds’, and drag of bedding in the ‘Jenolan Beds’,
as the fault is approached. There is also the indirect
evidence of differences in metamorphic grade, the
‘Jenolan Beds’ having a noticeable lower grade,
siltstones west of the fault giving way to phyllites on
the east. House (1988) also suggested that the fault
post-dates the Jenolan granite intrusion (Hellgate
Granite herein), as the contact metamorphism evident
in the ‘Eastern Beds’ in Pheasant Creek adjacent to
the fault is missing from the ‘Jenolan Beds’.

Evidence for the low angle Bulls Creek
Thrust of House (Figs 4, 8) is given by the sharp
low-angle boundary separating probably older
siliceous sandstone and siltstone from outcrops of
the limestone and nearby quartz porphyry within
the Inspiration Point Formation. This boundary is
marked by shearing of the beds, brecciation of quartz
blocks, and considerable slickensiding. The evidence
suggests thrusting from the southeast with the folding
plunging shallowly to the north.

Shannon (1976) showed five faults cutting across
the limestone belt in McKeowns Valley. The three
southern ones, two south of and one north of Dillon’s
Creek are parallel, trending NNE, with the southern
sides displaced easterly a small distance. The two
more northerly faults, north and south of Hennings
Creek, trend SSE. However, all five appear to have
little regional significance as no displacement has

120

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

been recognized in the adjacent beds, either on the
west or the east. These faults were reproduced on the
geological map in Kelly and Knight (1993).

GEOMORPHOLOGY

Jenolan Caves is situated at an altitude of 790
m in the deeply-incised east-trending valley of the
Jenolan River. The sides of the valley are marked by
several prominent benches in the landscape. Although
partly caused by lithological variations these benches
are almost certainly old erosion surfaces (Kieman
1988, Osborne 1987), suggestive of valley- in- valley
formation, indicating episodic uplifts following long
periods of stability and slow down-cutting.

The Jenolan River valley is located at the
southern edge of a slightly undulating plateau, named
the Jenolan Plateau by Craft (1928), which is a partly-
exhumed, gently-domed surface of Late Palaeozoic
age revealed by the partial removal of a thin cover
of Permian glacial and fluvio-glacial and (possibly)
Triassic rocks (Branagan 1983). Craft (1928) gave
considerable latitude to the definition of the plateau,
writing that it “extends vertically from 3700 feet (1 125
m) to 4400 feet (1338 m) above sea level (the highest
point is Mount Bindo, 1359 m), with an average
elevation slightly greater than 4000 feet (1216 m)”.
The surface is generally fairly even, and extends at
the higher level westerly to Oberon. This high level
continues extensively south and southwest (as the
Boyd Plateau) from Jenolan, but northeast it is less
extensive, the surface here with elevation above 900
m consisting of Warlock Ridge, the narrow easterly
trending Black Range ridge, and further north Mini
Mini Range, with Gibraltar Rocks (1070 m) at its
easterly culmination.

The Caves area is drained by the Jenolan River,
which commences in McKeowns Valley on the
west side of the Jenolan Caves Limestone, flowing
southerly and controlled by the strike of the limestone
and associated beds, then continuing underground
through the limestone belt before emerging on the
east side of the Grand Arch. Then it flows easterly,
possibly structurally controlled by recumbent folding
plunging towards the south and north (Kiernan 1988),
through Hellgate Gorge, then north-easterly to join
the Cox River at a ‘concordant’ junction, indicating
perhaps that the Jenolan River is a long-established
part of the Cox River system. Taylor (1958:145),
possibly following Sussmilch (1911:40), suggested
that water from McKeowns Valley flowed through
the caves system, at five different levels at different
times in its history, marking possibly five separate

phases of erosion (down-cutting) in the formation
of the Jenolan Valley. While the uppermost reach of
the Jenolan River (McKeowns Valley) has a course
largely controlled by the structural trend of the Jenolan
Caves Limestone, its swing across the limestone and
consequent eastern flow are oblique to the geological
‘grain’, and may represent superposition of an old
course on an uplifted surface.

The eastern slope of the Jenolan valley is drained
by the south-flowing Bulls Creek (the main tributary,
nearly 8 km long, of the Jenolan River), which heads
several km east from the north end of Binoomea
Ridge. The lower reaches of this stream contain some
alluvial terraces where flow is intermittent, and the
valley floor is relatively wide (House 1988).

The region east of Jenolan has been lowered
by the action of the long-established Cox River and
its numerous small tributaries. The Cox River has a
complex pattern, beginning in a shallow, broad valley
in the vicinity of Blackmans Flat (near Lidsdale),
in Permian rocks, then cutting deeply (structurally
controlled) through Late Devonian quartzites west of
Mount Walker (near Marrangaroo), flowing south of
Wallerawang in deeply-weathered rock (part of the
Bathurst-Hartley granite intrusion), now flooded by
construction of the Lyell Dam (Howes and Forster
1997). From near Lawson’s Sugarloaf, about 4 km
upstream from the junction of the Cox and the River
Lett, the main stream of the Cox River follows a
meandering course in a fairly broad valley for about
38 km, dropping steadily over a distance of about
19 km from an altitude of about 760 m near the Old
Bowenfels-Rydal Road to 600 m five km north of the
Cox River Rd-Lowther Road. It subsequently proceeds
another 19 km through an increasingly narrow valley,
decreasing to 510 m west of Megalong; then over
only 7 km, dropping to 304 m (near Pinnacle Ridge
on the east and Gibraltar Rocks on the west), then
to 150 m. The level of the Cox River then declines
very slowly over more than 30 km to well beyond its
junction with the Kowmung River.

We disagree with Craft (1928) who believed
that the older Palaeozoic rocks were less resistant to
erosion than the adjacent beds of the Sydney Basin
Permian-Triassic succession, and that they were worn
down relatively more rapidly. To explain the present
relationship between the higher Jenolan Plateau and
the Sydney Basin landscape, Craft suggested that the
Jenolan Plateau surface had been uplifted with ‘greater
recent elevation than the remainder of the surrounding
country’. However, there seems little reason to
explain the history of local landscape development
thus. Our field observations indicate that the Sydney
Basin sedimentation was restricted essentially to the

Proc. Linn. Soc. N.S.W., 136, 2014

121

GEOLOGY OF JENOLAN CAVES REGION

region presently covered by these rocks. The present
highest point on the Sydney Basin occurs a short
distance east of Cullen Bullen, at approximately 1280
m, whereas the Jenolan Plateau has numerous points
well above this elevation, as mentioned previously.
North from the Jenolan Plateau, Permian and Triassic
sedimentation of the Sydney Basin is restricted in
the Portland-Cullen Bullen-Ben Bullen region by
resistant ridges of Devonian and Silurian rocks,
and nearer to Jenolan, Sydney Basin sedimentation
is similarly restricted by Late Devonian quartz-rich
rocks at Mount Lambie to the west.

Scott and Pain (2006), based on work of the
BMR Palaeogeographical Group (1993), indicated
that the Jenolan Plateau is part of a much larger late
Palaeozoic erosion surface covering a wide area
of mid-western New South Wales and much of the
Lachlan Fold Belt region south in Victoria (see also
Blewett 2012:259, fig. 5.5). Examples of this ancient
landscape can be clearly seen east of Mudgee, near
Ben Bullen, and at the western edge of the Capertee
Valley where the surface on which the sediments of the
Sydney Basin began to be deposited is clearly dipping
easterly, tilted by late Carboniferous movement. The
existence of this old erosion surface accords with the
now generally accepted idea, based on considerable
evidence, that much of Australia’s landscape is old
and that modification has been slow (Young 1983,
Bishop 1985, Gale 1992, Twidale and Campbell
1993). However, contradictory evidence based on
apatite fission-track thermo-chronology (Blewett
2012:261, fig. 5.23) suggests considerable denudation
(up to 4 km) over vast areas of Australia, including
much of the supposed long-exposed landscape. These
contradictions provide a major problem which, at
present, shows little sign of resolution.

The Bathurst- Hartley-Jenolan Granite problem

An important event in the geological history
of the region was the post-orogenic intrusion of the
Kanimblan age Bathurst-Hartley granite body and
associated smaller intrusions, such as that cropping
out on the Jenolan River east of the caves (Hellgate
Gorge). While the granite intrusions took place after
the folding of the early-mid Palaeozoic Lachlan Fold
Belt rocks, there is little evidence of the depth at which
the intrusion was emplaced. Timing of the unroofing of
this body is a key element in the understanding of the
geomorphological history of the region, particularly
given the suggestion by Osborne et al. (2006) that
some cave sediments are of early Carboniferous age,
and that the Jenolan region must have been essentially
uncovered during the Carboniferous.

Vallance (1969) discussed the geology of the

Bathurst (and associated) intrusions, dealing with its
petrological variations and notably those mapped by
Joplin at Hartley (Joplin, 1931, 1933, 1935, 1944), and
suggested that the cover at the eastern end of the main
igneous body was ‘not more than 1500 m’. According
to Vallance (1969) the granite had not been deeply
eroded, although Howes and Forster (1997) indicate
that weathering at the Lyell damsite was greater than
expected.

Assuming this interpretation of cover thickness is
correct, erosion of the material capping the Bathurst
granite must have been very rapid, assuming that
it took place essentially during Carboniferous to
earliest Permian time. This leads directly into another
important question: where were these considerable
quantities of eroded sediments redeposited?
Sussmilch (1911:38) pointed out that a cutting south
of Lowther consisted of Permian conglomerate
containing ‘large water- worn boulders of quartzite
and granite imbedded in a matrix of granite detritus
(arkose), the whole resting upon an eroded granite
surface’. Sussmilch recognised the conglomerate
as belonging to what is now called the Shoalhaven
Group, the basal Permian unit of the Sydney Basin
succession. We now accept this unit as being, at least
in part, of glacial or fluvio-glacial in origin, a matter
that was not considered by the earlier workers. So
erosion, possibly with some reworking, had clearly
unroofed much of the granite by early Permian time,
involving removal of possibly 1500 m of cover over
an interval of some 50 million years. However, there
is little record of the deposition of Carboniferous
sediments adjacent to the zone of suggested erosion.
The nearest evidence of late Carboniferous-earliest
Permian deposition is found at the south-western edge
of the Sydney Basin, where Herbert (1980) delineated
a fluvioglacial drainage pattern (Talaterang Group
- see also Tye et al. 1996) and a northerly ‘tributary’,
the Burrawang Conglomerate, largely buried beneath
younger Sydney Basin sedimentary rocks, and whose
north-western extent is uncertain.

Modification of the Jenolan landscape clearly
continued though Permian and Mesozoic time, with
eroded material contributing to the formation of the
Sydney Basin, although most evidence indicates that
the bulk of that sedimentation came from the north and
south. There is little evidence of an easterly-flowing
drainage pattern which contributed to such erosion
and consequent deposition. The relatively recent
modifications of the Jenolan region in the Cenozoic
are the result of differing surface weathering with
the development of a variable regolith, and erosion,
mainly by the Cox River and its tributaries (draining
south and then east), and the Fish River and its

122

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

tributaries on the west, draining northerly and then
west and northwest in the Macquarie system.

While much of the above discussion is
speculative, it seems appropriate to draw attention
to these questions, which have not been previously
addressed, but which impinge on our understanding
of landscape evolution in the region.

Minor landscape-forming events and features

The Jenolan Fault has significant topographic
expression between Mount Inspiration and the
northern end of Pheasants Nest Creek, controlling
saddle development. Further north it has little effect
on topography, probably the result of relatively recent
exhumation from beneath Permian conglomerates.

Scree slopes, slumps and rockslides are common
throughout the region, but particularly in the eastern
area. The surface below Mount Inspiration has
prominent scarps with toes of slumps, consisting of
jumbled masses of blocks and boulders, which cover
the bedrock. House (1988) recorded a recent slump
which included a much greater proportion of fine-
grained material, and which clearly moved as a fluid.

Block streams with blocky, angular fragments
(ranging up to two m in maximum dimension) of
altered mafic volcanics and hornblende-bearing
dolerites that occur on the steeper slopes (up to 45°)
of gullies flanking Bulls Creek have been noted
particularly by Hallett (1988). The streams are
narrow, less than 20 m wide, and about 200 m long.
In the northwestern part of the area, Stewart (1987)
identified a series of block streams on eastern ridges
high above the river and not reaching it, whereas
debris flows and outwash fans were mapped at river
level along the Jenolan River (McKeowns Valley),
south of Navies Creek, the majority coming from the
eastern side of the valley (Fig. 4).

Consolidated gravels occur at various levels.
While some of these deposits are clearly related to
relatively recent changes in the presently established
streams, others, including some resting at high points
on the Jenolan Caves Limestone, may represent events
as far back as the Permian (Osborne and Branagan
1985). Consequently there is a considerable variability
in outcrop, and accessibility to the ‘solid’ rock. The
upper reaches of the tributary streams of the Jenolan
River generally show considerable outcrop, but the
lower reaches do not. Hill slopes are quite variable,
depending in part on the rock type, the more resistant
silica-rich units naturally being better exposed, but
siltstones often show surprisingly extensive outcrops.
In places the region is thickly vegetated, creeks are
very steep, often with waterfalls, and talus often
obscures outcrops, but there are some exceptions,

as noted by Hallett (1988), who suggested that some
rather smooth creek valley cross-sections indicated the
preservation of Permian valleys, possibly developed
through glacial or fluvioglacial processes.

GEOLOGICAL EVOLUTION AND HISTORY OF
KARSTIFICATION

The Jenolan story can only be understood in
relation to the history of the much wider picture of the
Lachlan Fold Belt. In general terms we are looking
at an area that was the focus of the deposition of
sediments in a gradually shallowing (and stabilising)
marine environment from Silurian to ?Early Devonian
times, followed by another period of subsidence and
shallowing (largely shallow marine to terrestrial) in
Late Devonian time. Volcanic activity was a continuing
factor. Intrusion of granite followed with some strong
earth movements, and the region underwent erosion
until early to mid-Permian time when the region was
subjected to glacial or peri-glacial conditions, and
sediments were deposited at the edge of a shallow sea
that deepened to the east.

Until the 1980s the age of karstification at
Jenolan and most other eastern Australian karst
was quite dogmatically stated as Quaternary, or at
the oldest, Pleistocene, post-dating the so-called
Kosciusko Uplift in Pliocene time. This is the
heritage of E.C. Andrews (1911). Ideas on the age of
formation of karst have been very strongly influenced
by Andrews’s Kosciusko Uplift hypothesis, which
became the revealed truth or dogma of Australian
geomorphologists until the 1970s. Andrews brought
the idea of very recent uplift, peneplanation and
erosional activity back to Australia after a visit to
America in 1908, where he was strongly influenced
by G.K. Gilbert and W.M. Davis. To some extent
these ideas of recent activity were also held by J.N.
Jennings, the result of his European experience, and
his influence among Australian speleologists here
was considerable during the 1960s-80s. It is probable
that Jennings was modifying these ideas before his
untimely death in 1984.

Karstification may have occurred during three
main periods: Middle Devonian, late Carboniferous-
early Permian and post Triassic. While modifications
to the cave system have occurred since Tertiary times,
the major karstification probably occurred earlier.
The difficulties of terrain and outcrop mean that many
problems remain to be elucidated in this challenging
area.

The development of karst in eastern Australia has
been a specific study of Armstrong Osborne, and his

Proc. Linn. Soc. N.S.W., 136, 2014

123

GEOLOGY OF JENOLAN CAVES REGION

findings are set out in a number of papers published
over the past twenty years (Osborne 1987, 1991, 1993,
1994, 1995, 1999; Osborne and Branagan 1985, 1988;
Osborne et al. 2006). They are especially summarised
in Osborne (1999) and Osborne et al. (2006), in
which the complexity of the story is pointed out,
with evidence for exhumation of McKeowns Valley
post-Permian, and the presence also of Cenozoic
bone-bearing gravels and a variety of surface and
underground drainage paths of various ages (see
also Kelly 1988). As Osborne (1984) showed, and
reiterated (Osborne 1999:14) the Jenolan Caves ‘are
not the product of a single recent event during which a
single process operated, but, rather, are the product of
a number of different events, during which a variety
of processes operated’. These events took place over
a geologically significant period of time.

Constraints on age of cave development

Evidence that karst development has been
proceeding since Carboniferous or even Early
Devonian time was proposed by Osborne et al. (2006),
who obtained K-Ar isotopic ages on illites from cave
deposits from a range of localities at Jenolan, the
oldest being late Emsian (Early Devonian), with no
fewer than nine results providing early Carboniferous
isotopic ages in the range 357 - 335 Ma, and a further
three falling into the later Carboniferous (325 - 313
Ma). A single sample yielded a late Permian age
(258.7 Ma). These results must be viewed within the
context of the overall geological history of the area,
and it is here that we observe certain areas of conflict
which we outline below. For rapid reference, an
extract of Osborne’s data is provided in Table 1, with
dates revised according to the latest geological time
scale (Gradstein et al. 2012).

Tectonic constraints.

The Silurian succession at Jenolan has been
affected by three significant tectonic episodes:
the earliest Devonian Bowning Orogeny, the mid-
Devonian Tabberabberan Orogeny, and the early
Carboniferous Kanimblan Orogeny. The present
attitude of the Jenolan Caves Limestone and its
associated sediments represents the combined effect
of all three of these orogenies.

The Kanimblan Orogeny, the last tectonic
episode within the Lachlan Fold Belt, concluded
with the intrusion of the Bathurst Batholith. Timing
of both these events has been the subject of recent
study, with ages for various phases of the Batholith
interpreted to range between 340 Ma and 312 Ma
(Pogson and Watkins 1998). Intrusions forming
part of the Batholith crop out less than 2 km from

Jenolan; the emplacement of these coarse granitic
bodies implies a depth of cover of the order of 1 .5 km
(Vallance 1969).

The duration of movements related to the
Kanimblan Orogeny appears to have been remarkably
brief. Glen (2013:337 and fig. 3) indicated an age of
340 Ma with no stated range; it was likely confined to
a brief interval in the earliest Visean. It follows that
any cave deposits still in original attitude must be no
older than this. This means that it is improbable that
any cave sediments of either Devonian or earliest
Carboniferous age could be horizontal.

The isotopic ages quoted in Table 1 cluster
around the Tournaisian and Visean, either coeval with
the Kanimblan Orogeny or just before it. There is also
significant overlap with the isotopic ages determined
for phases of the Bathurst Batholith. As emplacement
of coarse-grained granitic bodies requires substantial
depth, we can only conclude that dates for cave
deposit formation falling within this period have to be
regarded with caution. Furthermore, the range of K-
Ar isotopic ages on illites determined by Osborne et
al. (2006) from individual samples was considerable.
The two most extreme cases (JIC1, DCH4) covered
intervals of 83.04 Ma (mid Visean to late Permian)
and 55.42 Ma (latest Emsian to mid Visean).

Caymanite.

Cave deposits identified by Osborne et al.
(2006:379) as caymanites, by analogy with marine
deposits from Caribbean occurrences, occur notably in
the Devils Coachhouse, apparently from the locality of
their sample DCH4 for which ages ranging from late
Early Devonian to the later early Carboniferous were
determined. These horizontally to sub-horizontally
bedded sediments include crinoid columnals that are
certain indicators of marine conditions.

Marine sediments of this attitude necessarily
post-date the Kanimblan Orogeny, for simple
geometric reasons. Since no marine deposits of
Carboniferous age are known from anywhere within
the Lachlan Fold Belt, and in any case the succession
now exposed in the Jenolan area was of the order
of 1.5 km below the surface, the Carboniferous age
suggested by dating from sample DCH4, is, to say
the least, extremely unlikely, and it quite improbable
that this sample could be as old as Early Devonian.
The sole interval since the Carboniferous during
which the Jenolan area was under marine conditions
is the Permian, by which time the Bathurst Batholith
had already been unroofed. This is evident in the
area around Hartley and to the south, where basal
Sydney Basin marine sediments referred to the Berry
Formation directly and extensively overlie granites

124

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

J-H

.CD

• rH

J-H

.CD

• rH

/ ^Tl

• rH

/ ^n

2 s

?
• rH

OIj

■§

J-H

.(D

• rH
0
o

J-H

.CD

• rH
0
o

J-H

.CD

• rH
0
o

J-H

.CD

• rH
0
o

J-H

.CD

• rH
0
o

J-H

.(D

• rH
0
o

J-H

-CD

• rH
0
o

J-H

.CD

• rH
0
o

• rH
0
o
>
(D

• ^H

• rH

■HH

-HH

■§

• rH

§
• rH

■>

pp.

• rH

¥
• rH

§
• rH

5
0

• rH

-d

• rH

-d

• rH

-d

• rH

-d

• rH

-d

• rH

♦ rH

¥
• rH

£H

• rH
^H

* ^

' cd

Ifl

<L>

C D
m

L-

• rH

-o

-d

• rH
2

• rH

'd'

^ N

• rH
>
O

§
• rH

• rH

• rH
£

1
O i

0 s

-5

• rH

-d

• rH

-d

• rH

-d

• rH

¥
• rH

• rH

• rH
^H
0

• rH

♦ rH

• rH

§
• rH

pp.

: 3

c«

J-H

.CD

• rH
rH

• rH

J-H

■o

r-P

IT)

-H

T— H

»n

■d

—

l— H

T— H

♦ rH

0
♦ ^

ffi

J-H

ffi

0J)

OX)

_©

>—*

• rH

H-H

+->

• rH

c«

• rH
^H

• rH
>
CD

J-H

• rH
>
(D

• rH

• rH
>
(D

d -

N-

ffi

c3
• ^

• rH

a *§

fid

o a

Proc. Linn. Soc. N.S.W., 136, 2014

125

Table 1. Data selected from Osborne et al. (2006) showing K-Ar isotopic age dates determined on illites separated from samples of clay and caymanite
within Jenolan Caves; samples from identical sites are colour-coded. The ages determined in the original paper have been recalibrated to the latest

international time scale (Gradstein et al. 2012).

GEOLOGY OF JENOLAN CAVES REGION

of the Batholith (Bryan et al. 1966). Consequently a
Permian age for the caymanites, equivalent to that of
the Berry Formation, remains the most probable.

Mineralisation.

Osborne (1999) has put forward the attractive
idea that cupolate caves result from preferential
erosion of zones of sulphide-mineralised limestone.
Such mineralisation requires a source, and a review
of potential sources can provide significant data
relevant to both timing and origin. Subjacent sources
are suggested by the close proximity of components
of the Bathurst Batholith, as noted above. To these
can be added older sources, almost contemporaneous
with the Jenolan Caves Limestone itself. The acid
tuff, first noted by Sussmilch and Stone (1915) as
“rhyolite porphyry”, includes pumiceous fragments
(Fig. 10), indicating an eruptive origin and the
relatively close proximity of acid volcanism. The
numerous porphyries of probable Late Silurian age
shown on Figures 3-6 offer a further potential source.
Mineralization associated with the spilites generally
includes copper, which would almost certainly result
in staining of limestone in the caves. As this has not
been observed, this latter source is discounted.

CONCLUSIONS

We have clarified the proliferation of informal
stratigraphic names throughout the Jenolan area,
confirming correlation of the majority of rocks west
of the Jenolan Caves Limestone with the lower part
of the lower Silurian to Lower Devonian Campbells
Group. The age of the Jenolan Caves Limestone
(equivalent to the basal Mount Fairy Group) is revised
as mid-Wenlock. Conformably overlying the Jenolan
Caves Limestone to the east is a newly-defined unit,
the Inspiration Point Formation, characterized by
felsic volcanics and associated sedimentary rocks
that resemble the succession in the lower to mid
Mount Fairy Group (Fig. 3). The Inspiration Point
Formation includes limestone (previously referred to
as the ‘Eastern limestone’) that contains rare corals
and brachiopods of probable late Silurian age, thus
separating these outcrops from the Jenolan Caves
Limestone in both time and space.

With the exception of the younger units (Permian
conglomerates) that post-date the Kanimblan Orogeny
during the early Carboniferous and are essentially flat-
lying, the other stratigraphic units (in particular, the
steeply-dipping Jenolan Caves Limestone) become
progressively younger to the east, as determined by

bedding (often overturned).

Major faulting (e.g. McKeowans and Jenolan
faults) in the region mainly trends N-S. The type of
faulting is not always clear, but some strike-slip is
implied, as well as possible thrusting (Bulls Creek
Thrust). The faults separate the region into a number
of structurally-controlled domains, and tend to
obscure stratigraphic and depositional relationships
in a succession that is (apart from the Jenolan Caves
Limestone) generally devoid of fossil control.
However, some marker beds are recognized in the
Inspiration Point Formation, which assists in mapping
and correlation across faulted boundaries.

The present geomorphology of the region
probably evolved from late Carboniferous-early
Permian time, and the general plateau surface,
representing this feature, has been widely exhumed
around Jenolan. “Steps” in the deep valleys indicate
episodic periods of valley formation.

Cave formation may have occurred during at
least three main periods (Middle Devonian, late
Carboniferous-early Permian and post Triassic), but
the evidence for Devonian and Carboniferous periods
of karstification must be treated with caution. While
modifications to the cave system have occurred since
Tertiary times, the major karstification probably
occurred much earlier.

ACKNOWLEDGEMENTS

DFB gratefully acknowledges use of detailed field mapping,
associated laboratory studies and reports compiled by
T.L. Allan (1986), W. Stewart (1987), M. Hallett (1988)
and M. House (1988) (tragically killed November 1999
by an underground collapse at NorthParkes mine), (all
of University of Sydney), and D. Doughty (1994) (UTS)
in preparation of this paper. Access to these and other
university theses consulted in the preparation of this paper
was made possible by the cooperation of the School of
Geosciences, University of Sydney, the School of BEES
at UNSW, and the University of Technology, Sydney. We
appreciate useful discussions with K.J. Mills, E. Holland,
R.A.L. Osborne, and other members of the former Jenolan
Scientific Committee (1986-1998). Armstrong Osborne
and Harry Burkitt assisted JWP and IGP in sampling fossil
localities in limestone of the Inspiration Point Formation
that were crucial to determining the age of this unit. Yong
Yi Zhen (Australian Museum) kindly undertook SEM
photography of the conodonts and assisted with preparation
of Fig. 9. Jane and Larry Barron provided helpful discussion
of thin sections of the quartz porphyry east of the Jenolan
Caves Limestone and the photographs comprising Fig. 10.
Thesis maps and the regional geology compilation were
redrawn for this paper by Cheryl Hormann (Geological
Survey of NSW, Maitland). Bruce Welch provided editorial

126

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

advice in relation to optimum resolution of the figures.
We thank reviewers Ian Cooper and Dennis Pogson for
their perceptive and helpful comments on the manuscript.
Percival and Pickett publish with the approval of the
Executive Director, Resources and Energy NSW.

REFERENCES

Allan, T.L. (1986). Stratigraphy and Structure of
Palaeozoic Rocks in the Jenolan Caves Area,

N.S.W. B.Sc. (Hons.) Thesis, University of Sydney
(unpublished).

Andrews, E.C. (1911). Geographical Unity of Eastern
Australia in Late and Post Tertiary Time, with
applications to Biological Problems. Journal and
Proceedings of the Royal Society of New South Wales
44 , 420-480.

Bischoff, G.C.O. (1986). Early and Middle Silurian
conodonts from midwestern New South Wales.
Courier Forschungsinstitut Senckenberg 89 , 1-337.

Bishop, P. (1985). Southeast Australian late Mesozoic and
Cenozoic denudation rates: a test for late Tertiary
increases in continental denudation. Geology 13 , 479-
482.

Blewett, R.S. ed. (2012). ‘Shaping a Nation: a Geology of
Australia’. (Geoscience Australia and ANU E-Press:
Canberra).

Branagan, D.F. (1983). The Sydney Basin and its vanished
sequence. Journal of the Geological Society of
Australia 30 , 75-84.

Branagan, D.F., Holland, E. and Osborne, R.A.L. (1996).
The Geology of the Jenolan Region. In E. Hamilton-
Smith (ed .) Abstracts of Papers, Karst Studies
Seminar, Naracoorte, South Australia. Regolith
Mapping, Hamilton, Victoria, p. 17.

Branagan, D.F. and Packham, G.H. (1967). ‘Field Geology
of New South Wales’. 1 st edition. (Science Press,
Sydney).

Branagan, D.F. and Packham, G.H. (2000). ‘Field Geology
of New South Wales’. 3 rd edition. (Department of
Mineral Resources: Sydney).

Bryan, J.H., McElroy, C.T. and Rose, G. (1966). ‘Sydney
1:250,000 Geological Series, Explanatory Notes’.
(Geological Survey of New South Wales: Sydney).

Bureau of Mineral Resources Palaeogeographic Group
(1993). ‘Australia: Evolution of a continent’. (Bureau
of Mineral Resources, Canberra).

Came, J.E. (1908). ‘The Copper-Mining Industry and the
Distribution of Copper Ores in New South Wales’.
Mineral Resources No. 6 . (NSW Department of
Mines: Sydney).

Carne, J.E. and Jones, L.J. (1919). ‘The Limestone

Deposits of New South Wales’. Mineral Resources
No. 25. (NSW Department of Mines: Sydney).

Chalker, L. (1971). Limestone in the Jenolan Caves Area.
Records of the Geological Survey of New South Wales
13 ( 2 ), 53-60.

Chand, F. (1963). The geology of an area northeast of
Jenolan River. B.Sc. (Hons) Thesis, University of
NSW, Sydney (unpublished).

Cooper, I. (1990). Intrusive control of speleogenesis,
Mammoth Cave, Jenolan. SUSS Bulletin 30 ( 1 ), 21-
24.

Cooper, 1. (1993). Geology and speleogenesis, in caves of
Serpentine Bluff. SUSS Bulletin 33 ( 1 ), 28-30.

Craft, F.A. (1928). The Physiography of the Cox River
Basin. Proceedings of the Linnean Society of New
South Wales 53 ( 3 ), 207-254.

Curran, J.E. (1899). ‘Geology of Sydney and the Blue
Mountains’. 2 nd edition. (Angus & Robertson:
Sydney).

David, T.W.E. (1894). President’s Address, including a
Sketch of our present Knowledge of the Geological
history of Australia, Tasmania, and New Zealand,
from Archean Time down to the commencement of
the Permo-Carboniferous Period. Proceedings of the
Linnean Society of New South Wales 8 ( 4 ), 540-607.

David, T.W.E. (1897a). Anniversary Address: Summary
of our present knowledge of the structure and origin
of the Blue Mountains of New South Wales. Journal
and Proceedings of the Royal Society of New South
Wales 30 , 33-69.

David, T.W.E. (1897b). The occurrence of radiolaria in
Palaeozoic rocks in N.S.Wales. Proceedings of the
Linnean Society of New South Wales 21 ( 4 ), 553-570.

David, T.W.E. and Pittman, E F. (1899). On the Palaeozoic
radiolarian rocks of N.S.W. Quarterly Journal of the
Geological Society of London 55 ( 1 - 4 ), 16-37.

De Decker, P. (1976). Late Silurian (late Ludlovian)

conodonts from the Kildrummie Formation, south of
Rockley, New South Wales. Journal and Proceedings
of the Royal Society of New South Wales 109 , 59-69.

Doughty, D.G. (1994). The stratigraphy and structure of
the southern Limestone Area, Jenolan Caves, N.S.W.
B.Sc. Thesis, University of Technology, Sydney
(unpublished).

Etheridge, R., Jr. (1892). The Pentameridae of New South
Wales. Records of the Geological Survey of New
South Wales 3 ( 2 ), 49-60.

Etheridge, R., Jr. (1894). Appendix 6: Annual Report of the
Palaeontologist for the Year 1893. In Annual Report
of the Department of Mines and Agriculture, New
South Wales for 1893.

Fowler, T.J. and Iwata, K. (1995). Darriwilian-Gisbornian
conodonts from the Triangle Group, Triangle Creek
area. New South Wales. Australian Journal of Earth
Sciences 42 , 119-122.

Gale, S.J. (1992). Long-term landscape evolution in

Australia. Earth Surfaces Processes and Landforms
17 , 323-343.

Glen, R.A. (2013). Refining accretionary orogen models
for the Tasmanides of eastern Australia. Australian
Journal of Earth Sciences 60 , 315-370.

Gradstein, F.M., Ogg, J.G., Schmitz, M. and Ogg, G.
(2012). ‘The Geologic Time Scale 2012’. (Elsevier
Science Ltd: Oxford).

Proc. Linn. Soc. N.S.W., 136, 2014

127

GEOLOGY OF JENOLAN CAVES REGION

Gulson, B.L. (1963). The geology of an area southeast
of Jenolan Caves N.S W. B.App.Sc. Thesis,

NSW Institute of Technology [now University of
Technology], Sydney], (unpublished).

Hallett, M.S.C. (1988). The Geology of Bulls Creek
(North), Jenolan, N.S.W. B.Sc. (Hons.) Thesis,
University of Sydney (unpublished).

Havard, L.W. (1933). The Romance of Jenolan Caves.
Journal of the Royal Australian Historical Society.
[Limited facsimile edition. Department of Leisure,
Sport and Tourism, n.d.]

Herbert, C. (1980). 5. Southwestern Sydney Basin. In C.
Herbert and R. Helby (eds) ‘A Guide to the Sydney
Basin’. Geological Survey of New South Wales
Bulletin 26. Department of Mineral Resources, New
South Wales, 83 - 99.

House, M.J. (1988). The Geology of an Area centred
on Bulls Creek, Northeast of Jenolan Caves,

N.S.W. B.Sc. (Hons.) Thesis, University of Sydney
(unpublished).

Howes, M.S. and Forster, I. (1997). Engineering Geology
of Lyell Dam, LithgowNSW. In ‘Collected Case
Studies in Engineering Geology, Hydrogeology
and Environmental Geology. Third Series’ (Ed. G.
McNally) pp. 53-66. (Environmental, Engineering
and Hydrology Specialist Group, Geological Society
of Australia, Inc: Sydney).

Hunt, L., editor (1994). Jenolan Caves Reserve Future
Development Study: Final Report for Jenolan Caves
Reserve Trust. Colston, Budd, Hunt and Twiney Pty.
Ltd, August 1994, 120 pp. (unpublished).

Johns, R.K. ed. (1976). ‘History and Role of Government
Geological Surveys in Australia’. Ill pp.
(Government Printer of South Australia: Adelaide).

Joplin, G.A. (1931). Petrology of the Hartley District, i.
The Plutonic and Associated Rocks. Proceedings of
the Linnean Society of New South Wales 56, 16-59.

Joplin, G.A. (1933). Petrology of the Hartley District, ii.
The Metamorphosed Gabbros and Associated Hybrid
and Contaminated Rocks. Proceedings of the Linnean
Society of New South Wales 58, 125-158.

Joplin, G.A. (1935). Petrology of the Hartley District, iii.
The Contact Metamorphism of the Upper Devonian
(Lambian) Series. Proceedings of the Linnean Society
of New South Wales 60, 16-50.

Joplin, G.A. (1944). Petrology of the Hartley District,
v. Evidence of Hybridization in the Moyne Farm
Intrusion. Proceedings of the Linnean Society of New
South Wales 69, 129-138.

Kelly, B.F. (1988). The hydrogeology of the Northern
Limestone, Jenolan Caves and related land
management issues. B.Sc. (Hons.) Thesis, University
of NSW (unpublished).

Kelly, B.F. and Knight, M.J. (1993). Hydrogeology of
McKeown’s Valley, Jenolan Caves, New South
Wales, Australia. In ‘Collected Case Studies in
Engineering, Hydrogeology & Environmental
Geology, Second Series’. (Eds G. McNally, M.

Knight and R. Smith) pp. 72-91. (Environmental,

Engineering & Hydrogeology Specialist Group,
Geological Society of Australia, Inc: Sydney).

Kiernan, K. (1988). The geomorphology of the Jenolan
Caves area. Helictite 26(2), 6-21.

Link, A.G. and Druce, E.C. (1972). Ludlovian and

Gedinnian conodont stratigraphy of the Yass Basin,
New South Wales. Bulletin of the Bureau of Mineral
Resources, Geology & Geophysics 134, 1-136.

Lishmund, S.R., Dawood, A.D. and Langley, W V. (1986).
The Limestone Deposits of New South Wales.
Mineral Resources No. 25 (2 nd edition). 373 pp.
(Department of Mineral Resources: Sydney).

Macara, B. (1964). Petrology of the Budthingeroo
Metadiabase, a mafic body in the area of Jenolan
Caves. M.Sc. thesis, University of New South Wales,
Sydney (unpublished).

McClean, S.M. (1983). Geology and cave formation,
Jenolan Caves, N S.W. B.App.Sc. (geology) thesis,
NSW Institute of Technology [now University of
Technology], Sydney (unpublished).

Middleton, G.J. (1991). Oliver Trickett: Doyen of
Australia’s Cave Surveyors, 1847 - 1934. Sydney
Speleological Society Occasional Paper No. 10
[published in association with Jenolan Caves
Historical and Preservation Society], 156 pp.

Morrison, M. (1912). Notes on the geology of the country
between Rydal and Jenolan Caves. Annual Report,
Department of Mines, New South Wales for 1911,

189.

Osborne, R.A.L. (1984). Multiple karstification in
the Lachlan Fold Belt in New South Wales:
Reconnaissance evidence. Journal and Proceedings
of the Royal Society of New South Wales 117, 15-34.

Osborne, R.A.L. (1987). Multiple karstification: the

geological history of karsts and caves in New South
Wales, (including Case Study No. 3: Jenolan Caves).
Ph.D. Thesis, University of Sydney (unpublished).

Osborne, R.A.L. (1991). Palaeokarst Deposits at Jenolan
Caves, New South Wales. Journal and Proceedings
of the Royal Society of New South Wales 123(3-4),
59-74.

Osborne, R.A.L. (1993). Geological Note: Cave formation
by exhumation of Palaeozoic palaeokarst deposits at
Jenolan Caves, New South Wales. Australian Journal
of Earth Sciences 40, 591-593.

Osborne, R.A.L. (1994). Caves, dolomite, pyrite,

aragonite and gypsum: the karst legacy of the Sydney
and Tasmania Basins. Twenty-eighth Newcastle
Symposium on Advances in the Study of the Sydney
Basin. University of Newcastle (NSW, Australia),
Department of Geology Publication No. 606, 322-
344.

Osborne, R.A.L. (1995). Evidence for two phases of Late
Palaeozoic karstification, cave development and
sediment filling in south-eastern Australia. Cave and
Karst Science 22(1), 39-44.

Osborne, R.A.L. (1999). Presidential Address for 1998-
99. The Origin of Jenolan Caves: Elements of a New
Synthesis and Framework Chronology. Proceedings
of the Linnean Society of New South Wales 121, 1-27.

128

Proc. Linn. Soc. N.S.W., 136, 2014

D.F. BRANAGAN, J.W. PICKETT AND I.G. PERCIVAT

Osborne, R.A.L. and Branagan, D.F. (1985). Permian
palaeokarst at Billys Creek, New South Wales.
Journal and Proceedings of the Royal Society of New
South Wales 118, 105-111.

Osborne, R.A.L. and Branagan, D.F. (1988). Karst
Landscapes of New South Wales, Australia. Earth
Science Reviews 25, 467-80.

Osborne, R.A.L., Zwingmann, H., Pogson, R E. and

Colchester, D.M. (2006). Carboniferous clay deposits
from Jenolan Caves, New South Wales: implications
for timing of speleogenesis and regional geology.
Australian Journal of Earth Sciences 53, 377-405.

Packham, G.H. ed. (1969). The Geology ofNew South
Wales. Journal of the Geological Society of Australia
16, 1-654.

Pickett, J.W. (1969). Fossils from the Jenolan Caves
Limestone. Geological Survey ofNew South Wales
Report GS1969/519.

Pickett, J.W. (1970). Fossils from the Jenolan Caves

area. Geological Survey ofNew South Wales Report

GS1970/117

Pickett, J.W. (1981). Conodonts and the age of the Jenolan
Caves Limestone. Geological Survey ofNew South
Wales Report GS1981/305.

Pickett, J.W. ed. (1982). The Silurian System in New

South Wales. Geological Survey ofNew South Wales
Bulletin 29, 1-264.

Pickett, J.W. (2003). Corals and conodonts from the
Molong Limestone (Late Silurian), New South
Wales, Australia. [Abstract], Berichte des Institutes
fiir Geologie und Paldontologie der Karl-Franzens-
Univer sited Graz 7, 82.

Pogson, D.J. and Watkins, J.J. compilers (1998). Bathurst
1:250 000 Geological Sheet SI/55-8: Explanatory
Notes. 430 pp. (Geological Survey ofNew South
Wales: Sydney).

Pratt, R.T. (1965). Geology of the area between

Jenolan Caves and Ginkin, N S.W. B.Sc. (Lions)
thesis. University ofNew South Wales, Sydney
(unpublished).

Scheibner, E. and Basden, H. ed. (1998). ‘Geology of
New South Wales - Synthesis. Volume 2 Geological
Evolution. Geological Survey ofNew South Wales,
Memoir Geology 13(2). (Department of Mineral
Resources: Sydney).

Scott, K. and Pain, C.F. eds. (2008). Regolith Science’.
(C.S.l.R.O. Publishing: Melbourne).

Shannon, C.FI. (1976). Notes on Geology, Geomorphology
& Flydrology. In B.R. Welch (ed.) The Caves
of Jenolan 2: The Northern Limestone. Sydney
University Speleological Society, 5-9.

Shiels, O.J. (1959). Geology of the Tuglow River
area. B.Sc (Hons) thesis, University of Sydney
(unpublished).

Simpson, A. J. (1995). Silurian conodont biostratigraphy
in Australia: a review and critique. Courier
Forschungsinstitut Senckenberg 182, 325-345.

Simpson, C. (1986). Volcanic-plutonic associations within
the Bindook Volcanic Complex, Goodmans Ford
- Bullio area, New South Wales. Australian Journal
of Earth Sciences 33, 475-489.

Simpson, C.J., Carr, PF. and Jones, B.G. (1997). Evolution
of the Early Devonian Bindook Volcanic Complex,
Wollondilly Basin, Eastern Lachlan Fold Belt.
Australian Journal of Earth Sciences 44, 7 11-726.

Stanley, G.A.V. (1925). A preliminary account of
the geology and petrology of the Jenolan Caves
Reserve. B.Sc. (Hons) thesis. University of Sydney
(unpublished).

Stewart, W. (1987). Geology of an area north of Jenolan
Caves, N.S.W. B.Sc. (Hons) thesis. University of
Sydney (unpublished).

Strusz, D.L. (2007). The Silurian timescale - an

Australian perspective. Memoirs of the Association of
A ustralasian Palaeontologists 34, 157-171.

Strusz, D.L. and Munson, T.J. (1997). Coral assemblages
in the Silurian of eastern Australia: a rugosan
perspective. Boletln de la Real Sociedad Espahola de
Historia Natural, Seed on Geologica 92(1-4), 311-
323.

Sussmilch, C.A. (1911). A Geologist’s Notes on a Trip to
the Jenolan Caves. Technical Gazette ofNew South
Wales 1, 37-42.

Sussmilch, C.A. (1913). ‘Introduction to the geology of
New South Wales.’ (Angus and Robertson: Sydney).

Sussmilch, C.A. (1923). ‘Geological notes on the trip to
the Jenolan Caves. Guide-Book to the Excursion to
Blue Mountains, Jenolan Caves, and Lithgow’. (Pan
Pacific Congress: Sydney).

Sussmilch, C.A. and Stone, W.G. (1915). Geology of the
Jenolan Caves District. Journal and Proceedings of
the Royal Society ofNew South Wales 49, 332-384.

Talent, J.A., Berry, W.B.N. and Boucot, A.J. (1975).
Correlation of the Silurian rocks of Australia, New
Zealand and New Guinea. Geological Society of
America Special Paper 150, 1-108.

Talent, J.A., Mawson, R. and Simpson, A. (2003). Silurian
of Australia and New Guinea: Bio stratigraphic
correlations and palaeogeography. In E. Landing and
M.E. Johnson, eds, Silurian Lands and Seas. New
York State Museum Bulletin 493, 181-219.

Taylor, T.G. (1958). ‘Sydneyside Scenery’. (Angus &
Robertson: Sydney), 241 pp.

Thomas, O.D. and Pogson, D.J. (compilers) (2012).
Goulburn 1:250,000 Geological Sheet SI/55-12, 2 nd
edition, Explanatory Notes. Geological Survey of
New South Wales, Maitland.

Trickett, O. (1925). The Jenolan Caves (map). Jenolan
Caves Reserve Trust.

Twidale, C.R. and Campbell, E.M. (1993). ‘Australian
Landforms: Structure, Process and Time’.

(Gleneagles Publishing: Adelaide).

Tye, S.C., Fielding, C.R. and Jones, B.G. (1996).
Stratigraphy and sedimentology of the Permian
Talaterang and Shoalhaven Groups in the
southernmost Sydney Basin, New South Wales.
Australian Journal of Earth Sciences 43, 57-69.

Vallance, T.G. (1969). Plutonic and Metamorphic Rocks.

In G.H. Packham (ed.), Southern and Central
Highlands Fold Belt. Journal of the Geological
Society of Australia 16(1) 180-200.

Proc. Linn. Soc. N.S.W., 136, 2014

129

GEOLOGY OF JENOLAN CAVES REGION

Veevers, J.J. (2000). Time Scale. In J.J. Veevers (ed.),
‘Billion-year Earth History of Australia and
Neighbours in Gondwanaland’, pp 3-5. (GEMOC
Press: Sydney).

Walliser, O.H. (1964). Conodonten des Silurs.

Abhandlungen des Hessischen Landesamtes fur
Bodenforschung 41 , 1-106.

Welch, B.R. ed. (1976). ‘The Caves of Jenolan 2:

The Northern Limestone’. (Sydney University
Speleological Society: Sydney).

Wilkinson, C.S. (1884). ‘The Railway Guide of New
South Wales’ (Second Edition), pp 144-148. (New
South Wales Government Printer: Sydney).

Young, L. (1884). Description of the Jenolan Caves.

In Wilkinson, C.S. ‘The Railway Guide of New
South Wales’ (Second Edition). (New South Wales
Government Printer: Sydney).

Young, R.W. (1983). The tempo of Geomorphological
change: Evidence from Southeastern Australia.
Journal of Geology 91 , 221-230.

130

Proc. Linn. Soc. N.S.W., 136, 2014

A REVIEW OF THE CENOZOIC PALYNOSTRATIGRAPHY
OF THE RIVER VALLEYS IN CENTRAL AND WESTERN

NEW SOUTH WALES

Helene A. Martin

School of Biological, Environmental and Earth Sciences, University of New South Wales, Sydney

Australia 2052 (h.martin@unsw.edu.au)

Published on 10 June 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Martin, H.A. (2014). A review of the Cenozoic palynostratigraphy of the River Valleys in Central and
Western New South Wales. Proceedings of the Linnean Society of New South Wales 136 , 131-155.

The palynology of sediments from the Murray, Murrumbidgee, Lachlan, Macquarie and Namoi River
Valleys of the Western Slopes of New South Wales reveals remarkably similar patterns in the alluvium of
all of the valleys. Mid Miocene and older palynofloras found on the flood plains are rarely (if ever) seen in
the valleys where almost all of the palynofloras are placed in the late Miocene-Pliocene M. ga/eatus Zone. A
few palynofloras of the Pleistocene T. pleistocenicus Zone are found at the top of the sequence. The alluvial
fills of the palaeovalleys are similar also: in a basal late Miocene-Pliocene unit: the sands and gravels are
almost entirely quartz whereas the upper unit of Pleistocene age has a variety of resistant rock types and
only a minor quartz component. The alluvium of these river valleys is an important groundwater resource.

In the mid Miocene, a time of high sea level, the rivers of the Western Slopes discharged into the
flooded Murray Basin. Following major falls in sea level in the late Miocene, there was a basin-wide time
of erosion/non-deposition and entrenchment of the river valleys. Denudation associated with this regression
removed older sediments in the valleys and probably carved out the valley-in-valley structures. Tectonic
events were probably small and only maintained the elevation of the Highlands.

The palynofloras indicate a substantial change in the vegetation and climate over this time: from
rainforest and a wet climate in the mid Miocene to eucalypt sclerophyll forest and a drier, more seasonal
climate in the late Miocene-Pliocene to woodlands/grasslands and a much drier climate in the Pleistocene.
Deposition of the basal quartz rich alluvial unit occurred under a high rainfall, high-energy regime whereas
the upper unit was deposited under a drier climate and low energy regime.

Eustasy was a major forcing factor in the Neogene, but by Pleistocene time, the Murray Basin had
become isolated from the sea and the much drier climate had become the major forcing factor.

Manuscript received 23 October 2013, accepted for publication 19 February 2014.

KEYWORDS: climatic change, environmental history, eustasy, Neogene, palynostratigraphy, river valleys,
tectonics, western slopes of NSW.

INTRODUCTION

In the late 1950’s the then Water Conservation
and Irrigation Commission (now the NSW Office of
Water) began a drilling program to investigate the
groundwater potential in the Lachlan River Valley.
Prior to this time, most bores and wells were sunk for
stock water and domestic use and did not exceed 30
m in depth. Test drilling soon revealed good quality
water in much higher yields at greater depths, suitable
for irrigation and town water supply (Williamson,
1986). The program was extended to the other river
valleys of the Western Slopes (Fig. 1) and the valley
fills are an important groundwater resource.

This program required evidence from palyn-

ology for stratigraphic correlations, as only the
Cenozoic sand and gravels rather than the older
basement reliably yielded good quality water. Once
the palynology of the Lachlan River Valley was
established (Martin, 1987), similar patterns were
found in the palynology of the other valleys down
the Western Slopes of the Eastern Highlands (Martin,
1991), suggesting a similar geological history for all
the valleys. This study explores the geological and
environmental evolution of the valleys.

Today, the rivers of the Western Slopes drain into
the Murray Darling River System which discharges to
the sea at the mouth of the Murray in South Australia.
During most of the Cenozoic, however, they drained

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

1 * 2 * 144 * 146 ° 143 ° 150 ° 152 °

Figure 1. The stratigraphic palynology of the sediments has been studied in the areas as follows: 1.
Namoi and Gwydir River Valleys (Martin, 1980). 2. Namoi River Valley, Baan Baa to Boggabri (Mar-
tin, 1994). 3. Mooki River Valley, (Martin, 1979). 4. Spring Ridge District (1981a). 5. Castlereagh
River Valley (Martin, 1981b). 6. Macquarie River Valley (Martin, 1999). 7. Darling River (Martin,
1997). 8. Lake Menindee region (1988). 9. Murray Basin, Lachlan area (Martin, 1984b). 10. Lachlan
River Valley (Martin, 1987). 11. Murray Basin, Murrumbidgee area (Martin, 1984a). 12. Murray Ba-
sin, the Hay- Balranald-Wakool Districts (Martin, 1977). 13. Murray River (Martin, 1995).

into the Murray Basin that opened to the Southern
Ocean (the “Murravian Gulf’) (Fig. 2). The extent
of marine influence can be correlated with the global
supercycles of relative rise and fall of sea level (Brown
and Stephenson, 1991; Macphail et al., 1993).

The uplift of the Eastern Highlands and hence
formation of the Western Slopes has been a subject
of much debate. Most current hypotheses accept that
there has been little landscape evolution in many
regions since the early Cenozoic (e.g., Young and
McDougall, 1985; Veevers, 1991; Van der Beek et

al., 1999). Initial uplift has been attributed to isostatic
rebound due to erosional unloading or associated
with Cretaceous rifting of the eastern margin of the
continent (e.g., Webb et al., 1991). There have been
few claims of substantial uplift in the Cenozoic (e.g.
Holdgate et al., 2008) and this view has been contested
(e.g. Vandenberg, 2010). Most studies conclude that
further uplift during the Cenozoic has done little more
than maintain Highland elevation (e.g. Taylor et al.,
1983; Young and McDougall, 1985). Studies of the
Lachlan River Valley (Bishop and Brown, 1992) and

132

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

140 11 142 11 144 1 ' 146 *

Palaeoriver

A Ancestra! river

Fluvio-lacustrine

I I I I I

Marginal Marine

Marine

Figure 2. Early to mid Miocene paleogeography showing the marine incursion at its maximum extent in
the mid Miocene and the ancestral rivers from the Western Slopes flowing into the Murray Basin. After
Stephenson and Brown (1989).

Macquarie River Valley (Tomkins and Hesse, 2004)
infer uplift has occurred in the Neogene.

Interpretations of the paleovegetation and
climate indicate that both have changed considerably
over the course of the Cenozoic. The vegetation was
predominantly rainforest that required a wet climate
during the Palaeogene. In the Neogene, the vegetation
was predominantly sclerophyll forests that indicate a
drier climate. The mid to late Miocene was a time
of dramatic change (Martin, 1987, 2006; Macphail,
1997). All the evidence suggests that that there was a
precipitation gradient during the Cenozoic, parallel to
that of today, i.e. it was dryer to the west and wetter
to the southeast.

Eustasy, tectonics and climate have all had some
influence on the histories of the valleys. This study
attempts to evaluate the relative importance of each
factor through the Neogene.

METHODS

The NSW Office of Water (and its predecessors)
supplied sediment samples from bores. Most of the
samples were cuttings but some samples from cores
are included. Core samples were preferable but in

most cases, they were not available. The possibility
of contamination is greater with cuttings, both from
carry down with the circulating drilling mud and
from cavings, but with proper drilling and sampling
procedures, reliable samples may be obtained. For
investigative drilling, the mud is circulated until
it is clean of the coarse fraction and this greatly
reduces contamination. Additives to the mud are
not used (R.M. Williams, pers. comm.). If there
is contamination it can be detected, either in the
sediments themselves or in the preparations. A number
of bores penetrate both the Cenozoic and the older
basement and sampling across the boundary gives
some indication of contamination. Usually, there is
very little or no contamination unless sampling has
occurred very close to the contact. The large number
of barren samples interspersed with the polleniferous
samples would not be possible if contamination was a
problem (Martin, 1995).

The samples were first soaked in water then
treated with hydrochloric acid to remove all carbonates
if present. They were then treated with hydrofluoric
acid to remove silicates. These two acids together
removed all mineral matter. If sand and/or gravel
was present, it was removed by decanting early in

Proc. Linn. Soc. N.S.W., 136, 2014

133

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

MURRAY BASIN
Spore/pollen Zones
(Macphail, 1999)

Age

EPOCH

LACHLAN VALLEY
Spore/poiten 'phases'
(Martin, 1987)

T. pleistocenicus

PLEISTOCENE

7 f.

Asteraceae/Poaceae

M. lipsis

M. galeatus

PLIOCENE

Upper Myrtaceae

G'N

C. bifurcatus

- 10 m

Lower Myrtaceae
? ? ?

C. bellus

NEOCEN

MIOCENE

P. tuberculatus

- 20

77 n

- 25

Z j . u

- 30

OLIGOCENE

Upper N. aspersus

Mid N. aspersus

- 35 §
<

33.9

EOCENE

- 40

Figure 3. Summary of the palynological zones and the ages they
indicate. G/N, Gymnosperm and/or Nothofagus phase. The time
scale follows Ogg (2004) with the exception of the late Pliocene
Pleistocene boundary that follows Ogg and Pillans (2008). No-
menclature follows Macphail (1999). See text.

the treatment. Processing times and concentrations
varied with the nature of the sample. All treatments
were done with cold solutions.

The organic residues were oxidised with cold
Schultz solution (nitric acid saturated with potassium
perchlorate), usually with a 10% concentration for
10 minutes, but this stage was carefully controlled
according to the nature of the sample. The treatment
aimed to remove degraded organic matter that
obscured the pollen, but if too severe, it would also
destroy pollen. Treatment with an alkali (10% sodium
carbonate solution) removed the dark coloration,
making the samples suitable for examination under the
microscope. Again, times and concentrations varied,
depending on the nature of the sample. The oxidative
and/or alkali treatment may have been omitted with
samples that were naturally highly oxidised. Strew
samples of the residues were then mounted on a

microscope slide in glycerine jelly
(Martin, 1999).

Spore and pollen types were
identified according to descriptions in
Martin (1973a), Stover and Partridge
(1973), Macphail and Truswell (1989;
1993) and Macphail (1999) and were
counted along transects across the slide
to establish the relative abundance of
the common types. Testing showed
that a count of 120-140 grains was
a sufficient sample to represent the
quantitative aspects of the palynofloras.
The slides were then extensively
scanned for any uncommon types
missed in the count. The results were
used to assign the assemblage to a
palynological subdivision that could be
used for stratigraphy of the alluvial fill
of the valley.

The early work used palynological
subdivisions (‘phases’ in Martin,
1973b; 1987) based on quantitative
evidence in the Lachlan River Valley
for there was no published zonation of
the Neogene in southeastern Australia
that could be used. Inferred ages for
the ‘phases’ were attempted from the
geology in relation to basalts in the
region. Basalt was intersected in bores
1.5 km upstream from Eugowra on
Mandagery Creek, a tributary of the
Lachlan River. The mineralogy and
chemical composition was sufficiently
similar to basalt at Toogong, some 21
km further upstream, suggesting a common source for
both basalts (Williamson, 1986). The Toogong basalts
have been dated at 12.2 million years (Wellman and
MacDougall, 1974), or middle Miocene. More than
70 m of sediment above the basalt contained the
typical sequence of ‘phases’ found in the Lachlan
Valley and hence are upper Miocene and younger. Up
to 9 m of sediment below the basalt failed to yield
pollen (Williamson, 1986; Martin, 1987).

The subdivisions of Martin (1973; 1987) and
their inferred ages are listed below and in Fig. 3.
(Note: these inferred ages required testing but that
was not possible at that time. However, they served
the practical purpose of allowing some stratigraphic
control in these unconsolidated sediments that was
necessary for groundwater exploitation).

1 . The lower Myrtaceae phase of upper Miocene

134

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

age. Pollen of Myrtaceae is abundant and
Casuarinaceae may sometimes be common. Fern
spores may occasionally be abundant. Nothofagus
is not present or rare and the gymnosperm
content is usually low with Podocarpus the most
common type. Fem spores may be abundant in
some assemblages. A few rainforest angiosperms,
e.g. Quintiniapollis psilatispora ( Quintinia )
and Pseudowinter apollis (Winteraceae) may
be present also. This phase represents mainly
sclerophyll vegetation.

2. The Nothofagus phase of ?upper Miocene-
lower Pliocene age. The Nothofagus content
(Fuscospora and Lophozonia pollen types) is
relatively abundant. Rainforest angiosperms are
more common and there is a greater diversity of
gymnosperms.

3. The gymnosperm phase of ?upper Miocene-
lower Pliocene age may form a discrete entity
above the Nothofagus phase or may replace it
stratigraphically. The gymnosperms are more
diverse and include Dacrydium, Dacrycarpus
and Araucariaceae. The Nothofagus and
Gymnosperm phases represent more of the
rainforest element and may be useful for local
correlation.

4. The upper Myrtaceae phase of upper Pliocene
age is very similar to the lower Myrtaceae phase,
with the exception of the gradual disappearance
of rainforest pollen types, and an increase in
the Asteraceae (daisies) and Poaceae (grasses)
pollen towards the top of the sequence. If the
Nothofagus and Gymnospenn phases are not
present in the sequence, then the lower and upper
Myrtaceae phases cannot be separated. It also
represents sclerophyll vegetation.

5. The Asteraceae/Poaceae phase of Pleistocene
age. Asteraceae and Poaceae pollen become
abundant. Rainforest angiosperms and
gymnosperms are rare. This phase represents
woodlands and grasslands.

The palynofloras found in these phases are listed
in Tables 1 and 2.

A palynological zonation based on diagnostic
species for the Neogene was described for the Murray
Basin and published by Macphail and Truswell
(1993) and Macphail (1999). A similar palynological
zonation described for the Neogene of the Gippsland
Basin was published by Partridge (2006). These
zonations (Macphail, 1999; Partridge, 2006) are
based on diagnostic species and are considered more
reliable for correlation over large areas whereas the
system of Martin (1973b, 1987) may reflect local
ecological environments that can vary considerably

over large areas. The zonation for the Gippsland Basin
(Partridge, 2006) has been independently dated using
marine foraminiferal zonation and that for the Murray
Basin has been correlated with the Gippsland Basin
(Macphail, 1999). Zone equivalents of the Murray
Basin that are applicable to the river valleys and their
flood plains are as follows:

1 . The Middle Nothofagidites asperus Zone
Equivalent of upper Eocene age indicated by
the first appearance of Triorites magnificus
and Anacolosidites sectus. Proteacidites
rectomarginus and a diversity of Proteacidites
spp. are typical of the zone. Nothofagus
(. Nothofagidites spp.), especially the Brassospora
type dominates the palynofloras. The last
appearance of T. magnificus marks the top of the
zone.

2. The Proteacidites tuberculatus Zone Equivalent
of lower Oligocene to lower Miocene age.
Acaciapollenites miocenicus, Corsinipollenites
cf. C. epilobioides, Diporites aspis and Foveoletes
crater indicate the zone. Nothofagidites, the
Brassospora type dominates the assemblages
but Casuarinaceae {Haloragidites harrisii ),
Myrtaceae ( Myrtacidites) spp. or Phyllocladidites
mawsonii may occasionally be abundant.

3. The Canthiumidites (als Triporopollenites )
bellus) Zone Equivalent of upper lower Miocene
to middle Miocene age. The first appearance of T.
bellus and Symplocoipollenites austellus mark the
base of the zone. Haloragacidites haloragoides
and Rugulatisporites cowresis are also indicator
species. Nothofagus spp, Podocarpaceae and
Araucariaceae are the dominant pollen types.

4. The Monotocidites galeatus Zone Equivalent of
upper Miocene to lower Pliocene age (Macphail,
1999). The first appearance of M. galeatus
denotes the base of the zone. Myrtaceidites spp.
(Myrtaceae) and Casuarinaceae are the dominant
pollen types. There are two sub-divisions: the
Foraminisporis (als Cingulatisporites ) bifurcatus
of upper Miocene age and the Myrtaceidites lipsis
of early Pliocene age, each denoted by the first
appearance of their nominate species. Partridge
(2006) has elevated these two sub-zones to zones,
in place of the M. galeatus Zone.

5. The Tubulfioridites pleistocenicus Zone
Equivalent of upper Pliocene-Pleistocene
age (Macphail, 1999; Partridge, 2006). T.
pleistocenicus is consistently present and species
of Asteracae ( Tubulfioridites spp.) and Poaceae
(Graminidites media) become abundant.

The sequence of lower Mytaceae, Nothofagus ,

Proc. Linn. Soc. N.S.W., 136, 2014

135

Table 1. Palynofloras in Bores 14747 and 14505 of the Lachlan Formation (M. galeatus Zone Equivalent), from Martin (1969). Spore/pollen species are
described in Martin (1973a) and are expressed as percentages of total count. For location of bores, see Fig 5. For further distribution and botanical af-
finities of the spore/pollen species, see Martin (1987). Taxonomy follows Macphail (1999) where appropriate. Subtotals of important botanical groups are
given in bold for comparison with the Cowra Formation. Phases are as follows: 1, Upper Myrtaceae phase. 2, Gymnosperm phase. 3, Nothfagus phase.
4, Lower Myrtaceae phase.

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

l/o

I/O

l/o I/O
O

o o

(J\ <N

l/o

Tt-

l/o o

O <N

l/o

l/o O I/O o o

O O

l/o

o in ©
h o

in o
rd rd

in °

(S d d

o o

-r

in in o
<N O *-h‘

m o in ^

in
c\

CO CN
CO

in o o in in in
d d h d cn i

i/o in
in o

i/O

in in in
o o

S-H

0 <D

TO CO

cn
<1>
fa ^

C3 4^

<L> 3

£ <

S-H

1*0
CD

CD
CO CD

H -JR

•Ho

Oh ^

K u

S Oh

§ CD

u o

a -2
£ s;

8 <"R
CD

i*«^

H— >

o •

c+h o_;

u u

Oh
CL ^

» 1 8

1 1 §

^3 -d -d

o o o

gn 2 Oh

§ B >?

m q

$ o £

C-i-I Ch — 3 — J

u u u

c-w

O «
CO JD

i-H CO

H cn
cO cn
O <D

!-
Oh

S-H

• ^H

^Oh

.CD

• ^ Cu
CD CD
.Q CD
co

b c»
Q cn
CD <D

J-H

^ u

136

Proc. Linn. Soc. N.S.W., 136, 2014

Dacrydi urn (=Lygistipolleni tes) fiorini i Dacrydium spp. 1.0 2.5 1.5 5.0

Microcachryidites antarcticus Microcachrys 1.0 0.5 1.5 1.0

Podocarpus ( Dacrycarpites ) australiensis Dacrycarpus 6.0 7.5 0.5 12.5

H.A. MARTIN

^i-

-r

mi mi

in ©

^ ©
0> (N

o o

cd i

^h'

•

-r

•n

,—

' —

1—1

•n

•

,—l

•

^|-

•n

•

•n

«n

•

•n

«n

•

,—l

^h'

•n

^h'

•

-r

6
Q
Cj

<D <D
O O

1 >-H !—

<D &

C 3

c3
o o

cn cn

Xfl

;j w

^fflUUU

& <P

Q <D
OJ o
R c 3

£> 5-h

J 3

^3 O
o ir

Q ^

§P

J§

!>$

8 |
<D R
O R
c 3 O
O R

■c o

Oh
R Oh
m

R
a>

si c

s I

5h.

•

H-Hl

o .

<D Q
bJj Cj
O 3
PQ q

•

S Oh
hS

RS m
R c*2
~R 3

A ^
r s

t; r

R R
P .<3

a '■w

^ U

2 44

4)
i J
44

5s.

• ^
R
R
a

Oh
^ in
p O

Cj m

S-H

2 " <

s: m

Ps d

•-H (U

R So

00 cn

^ S3
a ?
to

~r »a

^ I

-a

A H <

Proc. Linn. Soc. N.S.W., 136, 2014

137

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

P (N

in in o

in »n o

'O 'O H

in in

in in in
©> ©> P

in in in in in

P P <n P ©

o in
P o

o in in in © in
P p © p <4 ©

in o

in in in in

O O O p

in ©

©>

in in ©
P o P

in in in
©> © in

in ©
in <n

©>

in ©

in in
© P

in in in

in in in © in

in © © in

in in in © ©
00 (N

©:

•

L-

•

CD
h— *

CD
h— >

CD
h— *

CD
h— >

S-H

■■

Ph .

• rH

S-H

©;

S-H

O) , ,

H— >

, ■ <

H— >

. P4

138

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

Table 2. Palynofloras found in the Asteraceae/Poaceae phase (=T. pleistocenicus Zone Equivalent) of the
Cowra Formation, from Martin (1969). Spore/pollen species are described in Martin (1973a) and are
expressed as percentages of total count. Taxonomy follows Macphail (1999) where appropriate. For
further distribution and botanical affinities of the spore/pollen species and location of bores, see Martin
(1987). Subtotals of important groups in the palynofloras are given in bold for comparison with the
Lachlan Formation.

Spore/pollen species

Bore

14578

12438

Depth (m)

17.3-17.8

14.3-15.5

19.2-20.1

SPORES

Nearest Living Relative

Lycopodium sp.

1.5

3.0

Deltoidospora inconspicua

?Adiantaceae

1.0

Cingulatisporites bifurcatus

Hepataicae

4.0

10.0

Reticularisporites (=Rugula-
tisporites) cowrensis

1.0

3.5

Total spores

GYMNOSPERMS

7.5

16.5

Podocarpus

(=Podocarpidites) elliptica

Podocarpus sens lat.

0.5

Total gymnosperms

ANGIOSPERMS :

0.5

DICOTYLEDONS

Acaciapollenites

myriosporites

Acacia

1.5

Casuarina ( =Casuarinidites )
• •

Casuarinaceae

0.5

1.0

0.5

camozoicus

Dodonaea sphaerica

Dodonaea spp.

0.5

Haloragacidites

haloragoides

Gonocarpus/Haloragis

1.0

Micrantheum spinyspora

Micrantheum spp.

2.0

Myrtaceidites eucalyptoides

Corymb ia spp.

1.0

8.5

1.0

M. mesonesus

Eucalyptus/Meterosideros

5.5

M. parvus

Myrtaceae

1.0

2.5

M. protrudiporens

Myrtaceae

3.0

2.0

Myrtaceidites spp. indet.

Myrtaceae

13.5

21.5

10.0

Total Myrtaceae

24.0

24.5

11.0

Onagraceae sp indet

Onagraceae

1.0

Polyporina bipatterna

0.5

P. ( =Chenopodipollis )
chenopodiaceoides

Chenopodiaceae

0.5

1.5

2.5

P. granulata

0.5

1.0

P. reticulatus

0.5

Polyporina spp.

4.5

0.5

2.5

Proteaceae cf. Grevillea

Grevillea

0.5

Asteraceae cf Cichoreae sp.
= Fenestrites

Asteraceae: Cichoreae

1.0

Tubulijioridites antipodica

Asteraceae

1.0

2.0

T. pleistocenicus

Asteraceae

13.5

5.5

45.5

T. simplis

Asteracee

23.0

15.5

18.5

Total Tubulijioridites spp.

ANGIOSPERMS:

MONOCOTYLEDONS

Asteraceae

37.5

24.0

64.0

Graminidites media

Poaceae

23.0

83.5

14.0

Lilia cidites sp.

1.5

Restionaceidites (=Milfordia)
hypolaeneoides

Restionaceae

1.5

Sparganiacepollis sp.

Sparganiaceae

0.5

Proc. Linn. Soc. N.S.W., 136, 2014

139

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

Gymnosperm and upper Myrtaceae phases of the river
valleys is thus equivalent to the M. galeatus Zone of
the Murray Basin (see Fig. 3). The diagnostic species
M. galeatus , C. bifurcatus and Dodonaea sphaerica
are common to both sequences and the general
quantitative aspects of abundant Myrtaceae and/or
Casuarinaceae are similar in both. However, there are
some notable differences: e.g., the diagnostic species
Myrtaceidites lipsis of the both the Murray and
Gippsland Basins has not been found in any of these
river valleys, and an equivalent of the Nothofagus
phase in the river valleys has not been reported
from the Murray Basin. These differences reflect the
environmental/ecological differences between the
marginal marine environments of the Murray and
Gippsland Basins and the totally non-marine and
upland environments of the river valleys.

The Asteraceae-Poaceae assemblage of the river
valleys is equivalent to the T. pleistocenicus Zone of
the Murray Basin. Both have the nominate species and
abundant Asteraceae (daisies) and Poaceae (grasses).

For this study, the Murray Basin zone equivalents
of upper lower Miocene to middle Miocene C.
bellus Zone, upper Miocene to lower Pliocene M.
galeatus Zone and the upper Pliocene to Pleistocene
T. pleistocenicus Zone are used. The upper Miocene-
lower Pliocene Nothofagus and/or gymnosperm
phase of the river valleys is retained as it has proved
to be a distinct local stratigraphic horizon. When
the recent changes to the Geologic Time Scale are
taken into account, viz. the recognition of the base
of the Quaternary at 2.6 Ma (Ogg and Pillans, 2008),
effectively incorporating the uppermost stage of
the Pliocene into the Pleistocene, the age of the M.
galeatus Zone becomes upper Miocene to Pliocene
and that of the T. pleistocenicus Zone becomes
Pleistocene (Fig. 3).

The palynology is presented in long profiles of
the valleys, with the bores adjusted for height above
sea level.

THE PALYNOSTRATIGRAPHY OF THE RIVER

VALLEYS

The Lachlan, Macquarie and Namoi River
Valleys have been the focus of investigations for they
have major groundwater potential and are considered
first. The Murray, Murrumbidgee, Castlereagh and
Darling River Valleys have not been investigated as
intensively but there is sufficient evidence to show
the overall patterns of alluvial deposition.

Lachlan River Valley

The Lachlan River catchment occupies an area
of about 90,000 km 2 . The river begins in the Great
Dividing Range and the headwaters arise at elevations
of up to 1,400 m at Mt. Canobolas. Most of the high
relief country is east of Cowra with only 2% classed as
rugged or mountainous. Alluvial flats of significance
commence about 13 km upstream of Cowra and 75%
of the catchment is classed as flat. Downstream of
Cowra, the alluvial flats become extensive and most
of the undulating landscape of the middle catchment
has been cleared (Williamson, 1986; Green et al.,
2011a)

The extensive flood plain environment of the
western part of the catchment is generally less than
200 m in elevation and features many wetlands
and effluent streams. Under nonnal conditions, the
Lachlan River is a terminal system with little water
flowing past the Great Cumbung Swamp at the end of
the river. Only in large flood events does water flow
into the Murrumbidgee River (Green et al., 2011a).

Test drilling reveals a buried ‘valley- in- valley’
structure downstream from Cowra to Jemalong Gap.
Remnants of an older valley floor are shown as a
shelf that maintains a depth of 27-30 m below present
drainage level but the depth of the valley carved
in the old valley increases markedly with distance
downstream. Williamson (1986) attribute this valley-
in- valley structure to successive tectonic movements
but suggests an alternative possible mechanism in the
effects of change in global sea levels (Williamson,
1986).

The alluvium in the Lachlan Valley is divided
into two distinct units: the basal Lachlan Formation
and the overlying Cowra Formation. The Lachlan
Formation consists of a series of interbedded
sediments ranging from gravels to clays. The sands
and gravels consist almost entirely of different
kinds of quartz and sometimes pebbles of chert but
they do not contain resistant rock types found in the
catchment. The clays may be divided into variegated
clays and carbonaceous clays: the latter are the
best for palynology. The sands and gravels of the
Lachlan Formation yield good quality water of low
salinity suitable for irrigation and town water supply
(Williamson, 1986).

The Cowra Formation disconformably overlies
the Lachlan Formation, i.e. there is an hiatus in
deposition between the two, and the strata range
from gravels to clays, all of which are predominantly
brown. The sands and gravels consist of the resistant
rock types found in the catchment and in this respect,
they differ significantly from the Lachlan Formation.
Carbonaceous clays are rare in the Cowra Formation.
The Cowra Formation yields water only suitable for

140

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

. 20

£D

01
tn>
■o

Euabalong
B'

Sea Level
elevated 100m

M. galeat us Zone
C. bellusZ one
P. tuberculatus Zone
Mid N. aspersusZone

Figure 4. Palynology of the Lachlan River, the alluvial plain section
(from Martin, 1987). For the ages the zones indicate, see Fig. 3.

stock and domestic purposes
(Williamson, 1986).

There is another unit found
in elevated positions and often
occurring as hill cappings: the
Glen Logan Gravels. This unit
consists of quartz gravel in a
red-brown silty matrix. It is
thought that they are remnants
of a formerly more widespread
unit that is stratigraphically
below the Lachlan Formation
and is probably the source of
the quartz sands and gravel in
the latter (Williamson, 1986).

Longitudinal sections
show the palynology of the
valley and alluvial plain and
onto the Murray Basin (Figs
4, 5). All of the palynofloras in
the valley fit the upper Miocene
to Pliocene M. galeatus
Zone, with the exception of
the upper lower Miocene to
mid Miocene T. bellus Zone
at the base of the Jemalong Gap bore. This bore is
exceptional, being located in the gap between the
Jemalong and Corridgery Ridges, the only feasible
gap where the ancient Lachlan River could go, and

Forbes

0 5 10 15 20 25

km L 20

Approximate scales m

has an exceptionally long sequence of carbonaceous
clays (Williamson, 1986). The base of Bore 30484
has an assemblage lacking diagnostic species and is
mainly Nothofagus, the Brassospora type that is more

Figure 5. Palynology of the Lachlan River Valley, the upstream section (from Martin, 1987). For the
ages the zones indicate, see Fig. 3.

Proc. Linn. Soc. N.S.W., 136, 2014

141

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

typical of the lower Oligocene to lower Miocene P.
tuberculatus Zone, but it could equally be an aberrant
T. bellus Zone. This bore is situated in the deepest
part of the alluvium on a former course of the river.
The sequence was first worked out for Bore 14747
(Fig. 5) where the Gymnosperm and Nothofagus
phases are distinct. Further downstream, these two
phases appear together. These two phases define a
useful local stratigraphic level.

The Pleistocene T. pleistocenicus Zone is not
found in the valley, except for the Jemalong Gap bore
where the uppermost polleniferous unit appears to
be intermediate between the M. galeatus Zone and
the T. pleistocenicus Zone. It occurs, however, in the
tributaries of the Lachlan River (Martin, 1987).

The T. bellus and M. galeatus Zones are found further
downstream of Jemalong Gap (Fig. 5). The sediments
of the Murray Basin are much deeper and the Middle
N. asperus Zone (upper Eocene) and P. tuberculatus
Zone (lower Oligocene to lower Miocene) are found
here.

Macquarie River

The Macquarie River originates in the Great
Dividing Range south of Bathurst and flows in a
northwesterly direction to join the Darling River
system near Brewarinna. The Macquarie-Bogan
Catchment covers an area of more than 74,000 km 2 .
(The Bogan River runs parallel to the Macquarie R. to
the southwest and the catchment between them is ill
defined. The Bogan River is an intermittent stream.)
Elevations across the catchment range from 1,300 m
in the mountains south of Bathurst to about 120 m near
Brewarrina in the northwest. BelowDubbo, the valley is
predominantly alluvial plain with an elevation of less
than 300 m (Middlemis et al., 1987; Green et al.,
2011b).

The valley consists of Palaeozoic Lachlan Fold
Belt rocks and Mesozoic sedimentary units deposited
in the Sydney-Gunnedah Basin. Basalts in the Dubbo
region range in age from 12.3 to 14.2 million years
(Ma). Sparse remnants of at least two widespread
Cenozoic depositional episodes are common in the
upper Macquarie area. These sediments are mainly
coarse grained and are found on the older, elevated
terraces (Smithson, 2010). A buried ‘valley- in-
valley’ form is seen between Wellington and Dubbo
(Tomkins and Hesse, 2004), similar to that in the
Lachlan Valley.

The sands and gravels of the basal Cenozoic
alluvium in the valley are predominantly quartz with
some chert and are interbedded with clays and organic
clay. The boundary with the overlying Quaternary
alluvium is usually distinct. The sands and gravels of

the Quaternary alluvium consist of variable lithologies
with a quartz content of only about 5% (Smithson,
2010). The clays and silts are mainly orange, red
and brown in colour. The Cenozoic alluvial fill of
the valley is thus very similar to that of the Lachlan
River Valley, and Middlemis et al. (1987) adopt the
names Lachlan and Cowra Formations (respectively)
for them, the names used by Williamson (1986) for
the equivalent sediments in the Lachlan River Valley.
The Quaternary alluvium is deposited on an erosion
surface (Tompkins and Hesse, 2004).

A longitudinal section of the valley (Fig. 6) shows
the palynology. Only the upper Miocene to Pliocene
M. galeatus Zone is found upstream of Narromine,
in the valley, with some occurrences to the south and
west of Narromine. Minor amounts of Nothofagus are
found in some bores, suggestive of the Nothofagus
phase. The upper lower Miocene to mid Miocene C.
bellus Zone is found south and west of Narromine,
in the alluvial plain and marks the course of former
channels. One occurrence of the lower Oligocene to
lower Miocene P. tuberculatus Zone is found west of
Narromine. Mesozoic basement assemblages have
been recorded from a number of the bores (Martin,
1999). The tributary Bell River (Fig. 7) has both the
upper Miocene to Pliocene M. galeatus Zone and
Pleistcene T. pleistocenicus Zone.

Namoi River Valley

The Namoi River catchment covers an area of
about 42,000 km 2 , from the Great Dividing Range
near Tamworth to the Barwon River near Walgett.
Elevations range from over 1,500 m in the south and
east to about 130 m on the alluvial floodplain in the
lower catchment west of Narrabri. Major tributaries
of the Namoi River include Coxs Creek, the Mooki
River and others further upstream of Boggabri. On
the floodplain west of Narrabri, where the river has
a low gradient, there is an increase in frequency of
lagoons and the development of several anabranches
and effluent streams (Green et al., 2011c).

The majority of the upper Namoi alluvium
overlies the sedimentary and volcanic rocks of the
Permian-Triassic Gunnedah Basin, the Jurassic
Oxley Basin sandstones and to the west, the Jurassic
Pilliga Sandstones of the Great Artesian Basin. The
alluvium of the Namoi River, the Coxs Creek and
Mooki River is divided into two layers: the shallower
Narrabri Formation and the deeper Gunnedah
Formation. The Narrabri Formation yields water
only suitable for stock. Good quality water suitable
for drinking can be found in aquifers across large
areas of the Gunnedah Formation and the highest
yields are found in the coarse sediments of the main

142

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

•£>

I I
I I
I I
I I

Narromine

Sea Level elevated 100m

T. pleistocenicus Zone

M. galeat us Zone
Gymnosperm/
Nothofagus phase
M. galeat us Zone

C. bellusZone
P. tuberculatusZone

i- o

L 20

Figure 6. Palynology of the Macquarie River Valley (from Martin, 1999). Section A- A’, the alluvial
plain. Section, B-B’, the river valley. For the ages the zones indicate, see Fig. 3.

palaeochannel that in most cases does not follow
the present drainage system (Barrett, 2012). On the
alluvial plain west of Narrabri, there is the older
Cubbaroo Formation underlying the Gunnedah

Formation. It was deposited in a pre-Cenozoic
channel following the northern limits of the alluvium
(Williams et al., 1989). This division of the alluvium
into three units in the lower Namoi is probably an

Proc. Linn. Soc. N.S.W., 136, 2014

143

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

Figure 7. Palynology of the Bell River, a tributary of the Macquarie River (Martin, unpubl.).

over-simplification (B. Kelly, pers. comm.) for the
alluvium has many aquifer zones that hold water with
widely varying salinities (Williams et al, 1989).

The palynology of the alluvial plain is presented
in the Appendix Fig. 8. Only the upper lower Miocene
to mid Miocene C. bellus Zone is found west of
Narrabri and it indicates a palaeochannel of the
former course of the river (Young et al., 2002). The
palynology shows that the Cenozoic sediments overly
an early Cretaceous basement. The palynofloras
found near Narraabri (Fig. 8) fit the upper Miocene
to Pliocene M. galeatus Zone but with an appreciable
Asteracea/Poaceae content.

The palynology of the Namoi River upstream
of Narrabri and of Coxs Creek (Fig 9) shows the
upper Miocene to Pliocene M. galeatus Zone with
abundant Myrtaceae and Casuarinaceae. The informal
Nothofagus phase, with relatively little Nothafagus ,
more abundant fern spores and gymnosperms is
identified in two bores in the same comparable
stratigaphic position as that of the Lachlan River Valley,
i.e. well down in the M. galeatus Zone. Araucariaceae
may be relatively abundant in the Nothofagus phase,
but also at a much shallower depth, near the top of the
M. galeatus Zone. The Pleistocene T. pleistocenicus
Zone may be found at shallow depths. Some bores
yielded basement Permian assemblages (Martin,
1994).

The Namoi River upstream of Gunnedah and the
Mooki River (Fig. 10) both yield the upper Miocene

to Pliocene M. galeatus Zone and the Pleistocene
T. pleistocenicus Zone, the latter at shallow depths.
The basement assemblages are Permian also (Martin,
1979). Narrow geological constrictions along the
length of the valley have had a significant affect on
how the alluvial sediments were deposited (Barrett,
2012 ).

Murray River Valley

The Murray River begins it course in the high
peaks of the Southern Alps of New South Wales and
Victoria. Altitudes range from about 2,200 m in the
east, to about 150 m at the Hume Dam near Albury.
The Upper Murray Catchment occupies about 1 5,330
km 2 and about one third of that is in New South Wales
(NSW Department of Primary Industries Office of
Water, 2013a). The Murray Riverina Catchment,
downstream of the Hume Dam covers 14,950 km 2
in southern New South Wales. It begins in the gentle
hills of the south western slopes where elevations
range from 300-600 m. Downstream of Corowa, the
river moves onto the flat plains of the Riverina where
elevations are less than 200 m (NSW Department of
Primary Industries Office of Water, 2013b).

The alluvial fill of the valley covers a basement
of Lachlan Fold Belt metamorphics and granites.
Downstream of Corowa, the Tertiary alluvium covers
the Lower Permian Oaklands-Coorabin coal measures
and further west, the river continues over the Murray
Basin (Martin, 1995).

The oldest sediments in the valley where the pre-

144

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

Narrabri fg

T. pleistocenicus Zone
/W. ga/eatws Zone
C. beliusZone

Early Cretaceous basement

r 0

L 20 m

(from palynology)

Bugilbone

INi

_ [eye I

elevated 100 m

Wee Waa

<rt

36287

ffi

Figure 8. Palynology of the Namoi alluvial plain (from Martin, 1980). For the ages the zones indicate, see Fig. 3.

Cenozoic basement is shallow are equivalent to
the late Miocene - Pliocene Lachlan Formation.
Downstream to the west of Corowa where these
sediments overlie those of the Murray Basin, they are
considered equivalent to the Calival Formation. The
sands of the Lachlan Formation are quartzose and
contain the main aquifers with only the upper part
containing rock fragments representative of the present
catchment rocks. The Shepparton Formation overlies
the Lachlan Formation and is characteristically
brown in colour. Quaternary sediments are assigned
to the Coonambigal Formation of the Murray Basin
(Martin, 1995).

The longitudinal section of the valley (Fig. 11)
shows the upper Miocene-Pliocene M. galeatus Zone
and the Pleistocene T. pleistocenicus Zone in the valley
upstream of Corowa. The M. galeatus Zone is also
found on the riverine plain downstream of Corowa.
The upper lower Miocene to mid Miocene C. bellus
Zone and the lower Oligocene to lower Miocene P.

tuberculatus Zone occur to the west and north of
Corowa (not shown on Fig. 8, Martin, 1995).

The upper Miocene-lower Pliocene Nothofagus
phase may be traced through the sequence in
stratigraphically the same relative position as in the
Lachlan River Valley alluvium, i.e. well down in the
M. galeatus Zone.

Murrumbidgee River Valley.

The Murrumbidgee catchment covers 84 km 2
in southern New South Wales. The river rises on the
Monaro plains at elevations of 2,200 m and flows
westwards to join the Murray River near Balranald,
where elevations are less than 50 m. A long narrow
flood plain extends upstream of Narrandera to the
foothills and yields good quality water suitable for
town water supply. Major irrigation areas are found
in the western part of the catchment (Green et al.,
201 Id)

In a small section across the valley at Narrandera,

Proc. Linn. Soc. N.S.W., 136, 2014

145

Coxs Creek

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

t'-

9frS9E

TV*9£

9£1?9€

5“

6D «
©

O Pi
a

Q\ ^
g <u

= £

fc o

-O

S-

^ 58 ^
6D aj
= 6Ii <U

58 o 5-
> PP U

M «

CS 5 J

J3 b
.33

O 5«
CtP <u

w fl
o

ON aj

O' J3

613

i 'fl

the upper part of the sediments is
considered the equivalent of the Cowra
and Lachlan Formations. The palynology
reveals the Pleistocene T. pleistocenicus
Zone, a T. pleistocenicus/M. galeatus
intergrade, the upper Miocene-Pliocene

M. galeatus Zone and the upper lower
to middle Miocene C. bellus Zone of
the Neogene sequence (Martin, 1973b).
The deep bores here penetrate sediments
of the Renmark Group of the Murray
Basin, and yield palynofloras of the
lower Oligocene to lower Miocene P.
tuberculatus Zone and the upper Eocene

N. asperus Zone that are extensive in
the Murray Basin further to the west
(Martin, 1984a; 1991). Narrandera is
situated on a long, narrow embayment
of the Murray Basin that is the earliest
recognisable stage of the Murrumbidgee
River System (Woolley, 1978).

Upstream at Wagga Wagga, the
sediments are considered the equivalent
of the Lachlan and Cowra Formations.
A small section across the river valley
has the upper Miocene-Pliocene M.
galeatus Zone and the Nothofagus phase
is particularly well represented with a
relatively high content of Nothofagus,
up to 27% of total count. The fern spore
count may be exceptionally high also,
50-80 % of total count but it is very
localised as another bore only 100 m
away did not yield high counts of spores
(Martin, 1973b; 1991).

Castlereagh River Valley

The Castlereagh River begins
in the Warrumbungle Ranges near
Coonabarabran and flows west to its
confluence with the Macquarie River.
The catchment has an area of 17,400
km 2 with elevations of 850 m in the east
to less than 200 m on the floodplains.
Stream flow is highly variable and
the sandy bed is often dry (NSW
Department of Primary Industries Office
of Water, 2013c). The Castlereagh River
is somewhat different to the other rivers
in that it was not a major tributary (J.
Ross, pers. comm.).

There are fewCainozoic palynofloras
in the Binnaway/Gilgandra/Curban part

146

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

fft

L 20 m

Breeza

Sea level elevated 200m

ffi

T. pleisto ceni cus Zone
M. galeatus Zone

Permian basement

(from palynology)

Figure 10. Palynology of the Mooki River Valley, tributary of the Namoi River (from Martin, 1979). For
the ages the zones indicate, see Fig. 3.

of the Castlereagh River Valley as basement is rather
irregular and palynofloras of Permian/Mesozoic age
are encountered at relatively shallow depths (Martin,
1981b). The upper lower to middle Miocene C. bellus
Zone is found around Gilgandra and downstream,
whereas the late Miocene-Pliocene M. galeatus Zone
occurs around Gilgandra (Martin, 1981b; 1991).

The upper part of the alluvium is consistently
brown, yellow, orange or reddish with minor grey
streaks or lenses. Consistently grey sediments are
encountered at deeper levels, but where palynofloras
are recovered, most of them are Mesozoic in age. It is
unclear if or how much of the sediments are equivalent
to the Lachlan Formation over this irregular basement
with so few Neogene palynofloras (Martin, 1981b).
Darling River

Palynology is available from only a few bores

along the Darling River, southwest of Bourke (Fig.
1). This part of the Darling does not flow down
the Western Slopes but follows an ancient fracture
lineament with a series of shallow grabens that
act as small basins (Martin, 1997). The Cenozoic
sediments form a linear belt along the lineament and
are divided into (1) an upper unit of grey silty clay of
the modern floodplain and probably the equivalent of
the Shepparton Formation in the Murray Basin, and
(2) a unit thought to be equivalent to the Renmark
Group of the Murray Basin. This latter unit consists
of sands and fine gravel, with carbonaceous muds at
the base (Martin, 1997).

The upper Eocene N. asperus Zone is found
at Tilpa, the lower Oligocene-lower Miocene P.
tuberculatus Zone at Glen Villa and upper lower to
middle Miocene C. bellus Zone at Jandra, the furthest
upstream. The Pleistocene T. pleistocenicus Zones

Proc. Linn. Soc. N.S.W., 136, 2014

147

Albury

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

9S£9£

s £

3 3
Ph «

I ®

=•«

e ©

3 .

S o 5 s

o> on

=3 ON

> .

3
s-

(Z3

1) Q

>3 *3

148

Proc. Linn. Soc. N.S.W., 136, 2014

373 m

H.A. MARTIN

occurs at shallow depths at Louth and Jandra (Martin,
1997).

West of Lake Menindee, a number of bores in an
area overlapping the edge of the Murray Basin (Fig. 1 )
only yielded palynofloras in dark grey clays at depths
greater than 70 m. Any grey clays at shallower depths
proved barren. It is thought that deep weathering
would have destroyed any pollen at shallower depths.
The deeper sediments would be equivalent to the
non-marine Renmark Group of the Murray Basin, but
dinoflagellates are commonly present in the southern
part of the area and indicate a marine environment
(Martin, 1988). The palynofloras indicate thick
sections of lower Oligocene-lower Miocene P.
tuberculatus Zone. This area would have been near
the edge of the upper Oligocene-mid Miocene marine
incursion when it was at its maximum extent in the
Murray Basin.

Further upstream in the Lower Balonne area near
St George, southern Queensland, correlatives of the
Lachlan Formation and Pleistocene palynofloras have
been identified (Macphail, 2004).

DISCUSSION

The palynological long profiles of the river
valleys reveal a striking similarity: accumulation of
the alluvial fill started in the upper Miocene in all
of the valleys, from the Murray in the south to the
Namoi in the north. Isolated occurrences of older
sediments in the alluvial fill are rare, although older
sediments of more than one age may be found in
elevated positions on the sides of the valleys. Older
sediments of the upper lower to middle Miocene and
the lower Oligocene to lower Miocene are almost
entirely restricted to the alluvial plains.

The lithology of the alluvium is also similar
in all the valleys. The sands and gravel in a basal
unit, corresponding to the upper Miocene-Pliocene
M. galeatus Zone, consist almost entirely of quartz
and yield good quality ground water. The overlying
unit, corresponding mainly to the Pleistocene T.
pleistocenicus Zone, contains a mixture of rock types
with quartz a minor component and the ground water
is of poorer quality.

The river valleys were in existence long before the
Neogene. A long, narrow embayment extending from
the Murray Basin into the highland area flanking the
Eocene plain to the west is the first recognisable stage
of the Murrumbidgee River. This embayment yielded
upper Eocene palynofloras at Narrandera. A similar
embayment appears to be present east of Hillston,
representing the earliest stages of the Lachlan River

(Wooley, 1978). The earliest identification of the
Murray River is Eocene in age (Macumber, 1978).

Paleogene palynofloras have not been found
in the river valleys of the Western Slopes, but both
Palaeogene and Neogene palynofloras may be found
throughout the Highlands and they are listed in Table
3. These palynofloras owe their existence to a basalt
cap that prevented subsequent removal by erosion.
Without any such protection, it is likely that any
Paleogene sediments in the valleys were removed by
an erosive event prior to the deposition of the Neogene
sequence. The upper Eocene and upper Oligocene-
lower Miocene sediments recovered from the small
basin-like structures along the Darling River suggest
that older sediments were deposited more widely but
palynofloras have only survived subsequent erosion
and weathering in these localised structures.

Tomkins and Hesse (2004) studied the Macquarie
Valley and found substantial vertical incision in the
mid-upper Miocene and interpreted it as evidence of a
single, high magnitude uplift event. They suggest that
first order tectonics were not synchronous with uplift
in the Lachlan Valley and that they were restricted
to relatively local spatial scales (Tomkins and Hesse,
2004). If first order tectonics were so different in the
mid-upper Miocene of the two adjacent catchments,
it is difficult to reconcile how the stratigraphy and
palynology of the alluvial fill came to be so similar
in both valleys.

A series of erosion terraces that are most
pronounced upstream of Cowra and diminish with
distance downstream are evidence of a series of
relatively minor uplifts (Williamson, 1 986). A study of
the Lachlan River Valley by Bishop and Brown ( 1 992)
concluded that Neogene isostatic rebound in response
to denudational unloading has been a significant
factor in maintaining highland elevation. Young and
McDougall (1985) studied the Eocene basalts of the
Shoalhaven valley. Post-basaltic denudation has been
slow and there has been little change in the landscape,
inferring very little uplift. Taylor et al. (1985) studied
the pre-basaltic topography of the northern Monaro
and concluded that there has been only minimal change
in topography and drainage during the Cenozoic,
suggesting no significant uplift. These studies thus
infer only relatively minor tectonics that would have
done little more than maintain the elevation of the
Highlands. Neogene studies of the southern part of
the Murray Basin indicate only minor tectonics that
eventually closed off the Murray Basin from the sea
(Wallace et al., 2005; McLaren et al., 2011).

The stratigraphy of the Murray Basin shows
a basin-wide erosion/non-deposition hiatus, the

Proc. Linn. Soc. N.S.W., 136, 2014

149

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

Table 3. Records of Palaeogene and Neogene Palynology in the Eastern Highlands

Locality

Palynological Zone/Age

Reference

Southern Monaro

Lygistepollenites balmei Zone, upper late
Palaeocene

Taylor et al. (1990)

Bowral area

L. balmei Zone, late Paleocene

McMinn (1989d)

Mt. Royal Range

L. balmei Zone, late Paleocene

Martin et al. (1987)

Nerriga

Malvacipollis Zone , early Eocene

Owen, (1975)

Bungonia

Lower N. asperus Zone, mid Eocene
Upper N. asperus Zone to lower P.

Truswell and Owen (1988)

Shoalhaven Catchment

tuberculatus Zone. Late Eocene-early
Oligocene

Nott and Owen (1992)

Invernell area

P. asperopolis Zone, mid Eocene and mid
P. tuberculatus Zone, late Oligocene

McMinn (1989e)

Glenn Innes

N. asperus Zone, mid and late Eocene

McMinn (1989a; 1989b;
1989 c)

Spring Ridge, Mooki R.

P. tuberculatus Zone, ?01igocene

Martin (1981a)

Cooma

Oligocene-late mid Miocene

Tulip et al. (1982)

Kiandra

Mid (?Late) P. tuberculatus Zone, early
Miocene

Owen (1975)

Cadia

C. bellus Zone, Mid Miocene

Owen (1975

Mudgee

C. bellus Zone, Mid to late Miocene

Martin (1999)

Gulgong area

C. bellus Zone, Mid to late Miocene

McMinn (1981)

Mologa Surface (Macumber, 1978), formed when
the sea retreated from the basin in the middle
early to late Miocene (~ 10 Ma) (Stephenson and
Brown, 1989). Macphail et al. (1993) has suggested
this unconformity correlates with the 13.8 or (the
preferred) 10.5 Ma eustatic sequence boundary of
Haq et al. (1987), when there were major falls in
global sea levels. Active entrenchment of adjacent
highland valleys also occurred at this time (Brown,
1989) when the older sediments in the valleys would
have been removed. The lowered base level may have
also carved out the valley-in- valley structure reported
for the Lachlan and Macquarie Valleys.

A study of Miocene eustasy off the northeastern
margin of Australia gives some measure of the late
Miocene fall in sea levels. There was a major drop of
53-69 m from 14.7-13.9 Ma (John et al., 2011). There
was another major fall in global sea level at 10.5 Ma
(Fig. 12), but the sediments were not suitable for an
estimation of the extent (John et al., 2011). However,
judging from the global sea levels of Haq et al. (1987,
Fig. 12), the total fall was probably about 200 m. This
latter drop corresponds to the time of the Mologa

erosional surface in the Murray Basin and active
entrenchment of the highland valleys (Brown, 1989).

The fall in sea level drained the Murray Basin and
the ancestral Murray River then flowed in a southerly
direction to discharge into the sea in western Victoria
(McLaren et al., 2011). This major marine regression
in the Murray Basin would have affected all of the
river valleys synchronously.

There was a short-lived marine transgression/
regression in the upper Miocene-Pliocene (~6
Ma, Fig. 12) (Brown, 1989). The rise in sea level
resulted in back filling of the previously excavated
entrenchments of the highland valleys (Stephenson
and Brown, 1989). Subsequently, there were only
minor fluctuations in sea level, restricted to the
southern part of the basin (Wallace et al., 2005;
McLaren et al., 2011). Relatively small amounts of
regional uplift defeated the drainage system and the
Murray Basin was cut off from the sea. A freshwater
megalake, Lake Bungunnia was formed -2.4 Ma, and
at this time, the rivers of the Murray Darling system
drained into L. Bungunnia. The ancestral Darling
River would have discharged into L. Bungunnia
about the Pooncarie-Mildura region, according to the
reconstructions of Stephenson and Brown (1989).

With the demise of Lake Bungunnia, the modern
course of the Murray River was established -700 ka

150

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

Figure 12. Global sea levels through the Neogene,
from Haq et al. (1987). A, the first major drop in sea
level -14.7-13.9 Ma. B, the second major drop in
sea level - 10.5 Ma (the time of the Mologa erosion/
non-deposition surface in the Murray Basin). C, the
short-lived marine transgression/ regression - 6 Ma.
See text for further explanation.

(McLaren et al, 2011).

Tomkins and Hesse (2004) reject the notion that
eustasy could have had an effect on the Macquarie
River Valley because of the distance to the coast of
over 1,500 km, but Miocene palaeogeography was
very different to that of today. The Macquarie River
joins the Darling River that would have discharged
into the sea about the Menindee region at the height
of the mid Miocene marine transgression (Fig. 2), a
much shorter course. Tomkins and Hesse (2004, p 285)

also describe “deposition in the upper Miocene-

lower Pliocene of the sediment demonstrates a

rising base level on the alluvial plain This rising

base level may have been caused by the short-lived
marine transgression/regression about 6 Ma that
resulted in the back filling of previously excavated
highland valleys (Stephenson and Brown, 1989). This
evidence suggests that eustasy from the mid Miocene

to Pliocene has had a considerable influence on the
histories of the valleys.

The Neogene was a time of changing climate
with decreasing precipitation (Macphail, 1997:
Martin, 2006). In the mid Miocene (T. bell ns Zone),
rainforest was widespread with precipitation of
> 1500 mm pa and relatively high humidity the
year round. By upper Miocene-Pliocene time (M
galeatus Zone), with mainly sclerophyll forest,
precipitation had decreased to -1500 -1000 mm
pa and there was a pronounced dry season and
fires occurred on a regular basis. In the short time
before the vegetation recovered from the fires, the
bare ground would have allowed increased erosion.
In the Pleistocene, with woodland/ grassland (T.
pleistocenicus Zone), rainfall had decreased further,
to < 1000 or probably 800 mm pa for the Lachlan
River Valley (Martin, 1987).

All the evidence suggests a rainfall gradient
parallel to that of today, i.e., it was dryer in the west
and wetter to the southeast. This gradient is seen in
the palynofloras and particularly in the Nothofagns
phase. The most Nothofagns in the palynofloras is
found in the Murray Valley with lesser amounts in
the Murrumbidgee and Lachlan Valleys. There is
also a gradient seen especially along the Lachlan
Valley, with more Nothofagns further upstream
and this gradient parallels the precipitation gradient
(Martin, 1987). It is thought the short-live marine
transgression/regression in the upper Miocene-lower
Pliocene (Brown, 1989) increased the precipitation
and allowed Nothofagns to migrate down the river
valleys from its refuge areas further up in the Eastern
Highlands. This resurgence, however, did not reach
the more westerly part of the slopes or the Murray
Basin, where, following the rainfall gradient, it
would have been drier than in the upper reaches of
the valley. (Note: Nothofagns is still present in a few
highland areas from the most south-easterly part of
Victoria to the Queensland border).

In all of the valleys, the change from the quartz
rich sands and gravels of the lower alluvial unit to
the variable lithologies with little quartz of the upper
unit is usually distinct and is described as an erosional
surface in the Lachlan and Macquarie Valleys
(Williamson, 1986; Tomkins and Hesse, 2004;
respectively). As far as the palynological method of
dating allows, it occurs about Pliocene-Pleistocene
time, but a probable cause is unclear. By this time, the
Murray Basin was cut off from the sea, hence isolated
from eustatic changes. However, the rivers drained
into the megalake Lake Bungunnia, formed about 2.4
Ma, under a climate with a much higher precipitation
than today. Lake levels fluctuated with climatic

Proc. Tinn. Soc. N.S.W., 136, 2014

151

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

fluctuations (Stephenson, 1986) that probably had
some influence on deposition/non-deposition in the
river valleys.

Williamson (1986) attributes the quartz in the
lower unit to reworking of a formerly widespread
older unit(s) whose remnants are found on elevated
parts of the valleys. The various lithologies in the
upper unit represent the resistant rock types of the
catchment. Tomkins and Hesse (2004) ascribe this
change in lithologies of the two units as a change in
the rock type being eroded. These two explanations
are not entirely satisfactory and they rely more on
more fortuitous events than anything else. There was
a marked climatic change about this time that would
have affected the whole of the Western Slopes. The
relatively high-energy depositional environment of
the lower unit would have decomposed more of the
less resistant rock types, leaving the resistant quartz.
With the change to decreased precipitation and a
lower energy environment in the Pleistocene, there
was less erosion and less chemical weathering, which
allowed more of the different rock types to survive
(Martin, 1987).

CONCLUSIONS

The palynology of all the river valleys of the
Western Slopes shows a remarkable similarity.
Palynofloras of the upper Miocene-Pliocene M.
galeatus Zone are the oldest found in the valleys
and occur in the basal quartz-rich sedimentary unit.
The overlying Pleistocene T. pleistocenicus Zone is
found in the upper sedimentary unit that has a mixture
of rock types and a minor quartz component. Older
sediments are found on the alluvial plains and in the
Highlands if those in the latter localities have been
protected from erosion by a basalt cap, but not in the
valleys.

Palaeogeography of western New South Wales
during the Neogene was very different to that of today.
The mid Miocene was a time of maximum marine
transgression in the Murray Basin and the rivers of the
Western Slopes drained into the Murray Basin. Major
falls in sea levels during the mid-upper Miocene
drained the Murray Basin and there was a basin-wide
erosion/non-deposition hiatus, with entrenchment in
the river valleys that would have removed the older
sediments. Subsequently, minor tectonics closed off
the mouth of the Murray Basin and the present course
of the Murray River is only -700 ka old.

Tectonics in the Highlands and Western Slopes
had a relatively minor and localised impact, probably

only maintaining the elevation of the Highlands.

The palynofloras indicate an increasingly drier climate
through the Neogene and into the Pleistocene, and
change from a high-energy to a low energy erosion/
deposition environment that would have influenced
the nature of the sediments being deposited.

Eustasy, tectonics and climate have all had some
influence on the histories of the river valleys. The
major changes in sea level through the Neogene,
however, would have impacted on all of the valleys,
synchronously, to produce the remarkable similarity
of the histories of the valleys. The major change to
a much drier climate in the Pleistocene would have
impacted on the whole of the Western Slopes, and
indeed far beyond the region of this study.

ACKNOWLEGEMENTS

1 am indebted to the New South Wales Office of
Water and its predecessors that have provided samples
from bores and financial support. Hydrogeologists in the
Office of Water have assisted with invaluable information
and support, particularly Ms Ann Smithson of the Dubbo
Office of Water. Dr Mike Macphail gave invaluable advice
about palynostratigraphy. A/Prof Bryce Kelly critically
commented on the manuscript. Mr Matthew Hunt prepared
the diagrams.

REFERENCES

Barrett, C. (2012). Upper Namoi Groundwater Source
- Status Report 2011. (NSW Department of Primary
Industries Office of Water, Sydney).

Bishop, P. and Brown, R. (1992). Denudational isostatic
rebound of intraplate highlands: the Lachlan River
Valley, Australia. Earth Surface Processes and
Landforms 17, 345-360.

Brown, C.M. (1989). Stmctural and stratigraphic

framework of groundwater occurrence and surface
discharge in the Murray Basin, southeastern
Australia. BMR Journal of Geology and Geophysics
11, 127-146.

Brown, C.M. and Stephenson, A. E. (1991). Geology of
the Murray Basin, Southeastern Australia. Bureau of
Mineral Resources, Geology and Geophysics Bulletin
235. 1-430

Green, D., Petrovic, J., Moss, P, Burrell, M. (2011a).

Water Resources and management overview: Lachlan
catchment pp 1-28. (NSW Office of Water, Sydney).

Green, D., Petrovic, J., Moss, P, Burrell, M. (2011b).
Water Resources and management overview:
Macquarie-Bogan Catchment pp 1-29. (NSW Office
of Water, Sydney).

Green, D., Petrovic, J., Moss, P, Burrell, M. (2011c).
Water Resources and management overview: Namoi

152

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

catchment pp 1-31. (NSW Office of Water, Sydney).

Green, D., Petrovic, J., Moss, R, Burrell, M. (201 Id).
Water Resources and management overview:
Murrumbidgee catchment pp 1-29. (NSW Office of
Water, Sydney).

Haq, B.U., Hardenbold, J. and Vail, P.R. (1987).
Chronology of fluctuating sea levels since the
Triassic. Science 235, 1156-1167.

Holdgate, G.M., Wallace, M.W., Gallagher, S.J., Wagstaff,
B.E. and Moore, D. (2008). No mountains to snow
on: major post-Eocene uplift of the East Victorian
Highlands; evidence from Cenozoic deposits.
Australian Journal of Earth Sciences 55, 211-234.

John, C M., Karner, G.D., Browning, E., Leckie, M R.
Mateo, Z. et al. (2011). Timing and magnitude of
Miocene eustasy derived from the mixed siliclastic-
carbonate stratigraphic record of the northeastern
Australian margin. Earth and Planetary Science
Letters 304, 455-487.

Macphail, M.K. (1997). Late Neogene climates in

Australia: fossil pollen and spore- based estimates in
retrospect and prospect. Australian Journal of Botany
45, 425-464.

Macphail, M.K. (1999). Palynostratigraphy of the Murray
Basin, inland southeastern Australia. Palynology 23,
197-240.

Macphail, M.K. (2004). Palynostratigraphic analysis of
core and cutting samples from the Lower Balonne
area, southern Queensland, Australia. CRC LEME
Open File Report 167, 1-40.

Macphail, M.K. and Truswell, E.M. (1989).

Palynostratigraphy of the central west Murray Basin.
BMR Journal of Australian Geology and Geophysics
11,301-331.

Macphail, M.K. and Truswell, E.M. (1993).

Palynostratigraphy of the Bookpurnong Beds and
related late Miocene-early Pliocene facies in the
central west Murray Basin, part 2: spores and
pollen. AGSO Journal of Australian Geology and
Geophysics 14, 383-409.

Macphail, M.K., Kellett, J.R., Rexilius, J.P and O’Rorke,
M.E. (1993). The “Geera Clay equivalent”: a
regressive marine unit in the Renmark Group that
sheds new light on the age of the Mologa weathering
surface in the Murray Basin. AGSO Journal of
Geology and Geophysics 14, 47-63.

Macumber, PG. (1978). Evolution of the Murray River
during the Tertiary period. Evidence from northern
Victoria. Proceedings of the Royal Society of Victoria
90, 43-52.

Martin, H.A. (1969). The palynology of some Tertiary
and later deposits in New South Wales. PhD thesis,
University of New South Wales (unpubl.).

Martin, H.A. (1973a). The palynology of some Tertiary
Pleistocene deposits, Lachlan River Valley,

New South Wales. Australian Journal of Botany
Supplement 6, 1-57.

Martin, H.A. (1973b). Upper Tertiary palynology in

southern New South Wales. Special Publications of
the Geological Society of Australia 4, 35-54.

Martin, H.A. (1977). The Tertiary stratigraphic palynology
of the Murray Basin in New South Wales. 1. The Hay-
Balranald-Wakool Districts. Journal and Proceedings
of the Royal Society of New South Wales 110, 41-47.

Martin, H.A. (1979). Stratigraphic palynology of the
Mooki Valley, New South Wales. Journal and
Proceedings of the Royal Society of New South Wales
112, 71-78.

Martin, H.A. (1980). Stratigraphic palynology from

shallow bores in the Namoi River and Gwydir River
Valleys, North-Central New South Wales. Journal
and Proceedings of the Royal Society of New South
Wales 113,81-87.

Martin, H.A. (1981a). An Early Cretaceous age for
subsurface Pillaga Sandstone in the Spring Ridge
District, Mooki Valley. Journal and Proceedings of
the Royal Society of New South Wales 114, 29-31.

Martin, H.A. (1981b). Stratigraphic palynology of the
Castlereagh River Valley, New South Wales. Journal
and Proceedings of the Royal Society of New South
Wales 114, 77-84.

Martin, H.A. (1984a). The stratigraphic palynology of
the Murray Basin in New South Wales. II. The
Murrumbidgee area. Journal and Proceedings of the
Royal Society of New South Wales 117, 35-44.

Martin, H.A. (1984b). The stratigraphic palynology on the
Murray Basin in New South Wales. III. The Lachlan
area. Journal and Proceedings of the Royal Society of
New South Wales 117, 45-51.

Martin, H.A. (1987). Cainozoic history of the vegetation
and climate of the Lachlan River region, New South
Wales. Proceedings of the Linnean Society of New
South Wales 109, 213-257.

Martin, H.A. (1988). Stratigraphic palynology of the Lake
Menindee region, northwest Murray Basin, New
South Wales. Journal and Proceedings of the Royal
Society of New South Wales 121, 1-9.

Martin, H.A. (1991). Tertiary stratigraphic palynology
and palaeoclimate of the inland river systems in New
South Wales. In The Cainozoic in Australia: a re-
appraisal of the evidence (Eds M.A ,J. Williams, P.

De Deckker and A.P. Kershaw). Special Publication
of the Geological Society of Australia 18, 181-194.

Martin, H.A. (1993). Middle Tertiary dinoflagellate and
spore/pollen bio stratigraphy and palaeoecology of the
Mallee Cliffs bore, central Murray Basin. Alcheringa
17,91-124.

Martin, H.A. (1994). The stratigraphic palynology of the
Namoi River Valley, Baan Baa to Boggabri, northern
New South Wales. Proceedings of the Linnean
Society of New South Wales 114, 45-58.

Martin, H.A. (1995). The stratigraphic palynology of
the Murray River Valley in New South Wales.
Proceedings of the Linnean Society of New South
Wales 115, 193-212.

Martin, H.A. (1997). The stratigraphic palynology of bores
along the Darling River, downstream from Bourke,
New South Wales. Proceedings of the Linnean
Society of New South Wales 118, 51-67.

Proc. Linn. Soc. N.S.W., 136, 2014

153

PALYNOSTRATIGRAPHY OF RIVER VALLEYS

Martin, H.A. (1999). The stratigraphic palynology of the
Macquarie River Valley, Central Western New South
Wales. Proceedings of the Linnean Society of New
South Wales 121, 113-127.

Martin, H.A. (2006). Cenozoic climatic change and the
development of the arid vegetation in Australia.
Journal of Arid Environments 66, 533-563.

Martin, H.A., Worral, L. and Chalson, J. (1987). The first
occurrence of the Paleocene Lygistepollenites balmei
Zone in the Eastern Highlands Region, New South
Wales. Australian Journal of Earth Sciences 34, 357-
365.

McLaren, S., Wallace, M.D., Gallagher, S .J., Miranda,

J.A., Holdgate, G.R., et al. (2011). Palaeogeographic,
climatic and tectonic change in southeastern
Australia: late Neogene evolution of the Murray
Basin. Quaternary Science Reviews 30, 1086-1111.

McMinn, A. (1981). A Miocene microflora from the
Home Rule kaolin deposit. Quarterly notes of the
Geological Survey, New South Wales 43, 1-4.

McMinn, A. (1989/a). Tertiary palynology of Glen innes
DDH 107. Geological Survey of New South Wales
Department of Mineral Resources Unpublished
Palynological Report 1989/1,

McMinn, A. (1989/b). Tertiary palynology of a sample
from Glen Innes DDH 110. Geological Survey of
New South Wales Department of Mineral Resources
Unpublished Palynological Report 1989/5.

McMinn, A. (1989/c). Eocene palynology of Glen Innes
DDH 105. Geological Survey of New South Wales
Department of Mineral Resources Unpublished
Palynological Report 1989/11.

McMinn, A. (1989/d). Early Tertiary palynology of
sediment from Rileys Sugarloaf, near Bowral.
Geological Survey of New South Wales Department
of Mineral Resources Unpublished Palynological
Report 1989/12.

McMinn, A. (1989e). Tertiary palynology in the Inverell
area. Quarterly Notes of the Geological Survey, New
South Wales 76, 2-10.

Middlemis, H., Lawson, S., Ross, J., Merrick, N. (1987).
Hydrology of the Lower Macquarie area, with
emphasis on the shallow alluvial groundwater
system, recent drilling and simple modelling results.
Department of Water Resources Technical Services
Report No.l, October 1987.

Nott, J.F. and Owen, J.A. (1992). An Oligocene

palynoflora from the middle Shoalhaven catchment in
the southeast Australian highlands. Palaeogeography,
Palaeoclimatology, Palaeoecology 95, 135-151.

NSW Department of Primary Industries Office of Water
(2013a). Upper Murray catchment, http ://www. water.
nsw.gov.au/ (Accessed June 2013).

NSW Department of Primary Industries Office of Water
(2013b). Murray Riverina catchment. http://www.
water.nsw.gov.au/ (Accessed June 2013).

NSW Department of Primary Industries Office of Water
(2013c). Castlereagh catchment, http ://www. water.
nsw.gov.au/ (Accessed June 2013).

Ogg, J.G. (2004). Status and divisions of the International
Geologic Time Scale. Lethaia 37, 183-199.

Ogg, J.G. and Pillans, B. (2008). Establishing Quaternary
as a formal international Period/System. Episodes 31,
230-233.

Owen, J.A. (1975). Palynology of some Tertiary deposits
in New South Wales. Ph.D thesis, Australian National
University (unpubl.)

Partridge, A.D. (2006). Late Cretaceous-Cenozoic

palynology zonations Gippsland Basin. In: Monteil,

E. (coord.). Australian Mesozoic and Cenozoic
Palynology Zonation - updated to the 2004 Geologic
Time Scale. Geoscience Australia Record 2006/2.

Smithson, A. (2010). Upper Macquarie Alluvium

- Groundwater Management Area 009, Groundwater
Status Report 2010. (NSW Office of Water, Sydney).

Stephenson, A.E. (1986). Lake Bungunnia - a Plio-
Pleistocene megalake in Southern Australia.
Palaeogeography, Palaeoclimatology, Palaeoecology
57, 137-156.

Stephenson, A.E. and Brown, C.M. (1989). The ancient
Murray River System. BMR Journal of Geology and
Geophysics 11, 387-395.

Stover, L.E. and Partridge, A.D. (1973). Tertiary and Late
Cretaceous spores and pollen from the Gippsland
Basin, Southeastern Australia. Royal Society of
Victoria Proceedings 85, 237-286.

Taylor, G., Taylor, G.R., Bink, M., Foudoulis. C . et al.
(1985). Pre-basaltic topography of the northern
Monaro and its implications. Australian Journal of
Earth Sciences 32, 65-71.

Taylor, G., Truswell, E.M., McQueen, K.G. and Brown,
M.C. (1990). Early Tertiary palaeogeography,
landform evolution and palaeochmates of
the Southern Monaro, N.S.W. Australia.
Palaeogeography, Palaeoclimatology, Palaeoecology
78, 109-134.

Tomkins, K.M. and Hesse, P.P. (2004). Evidence
of Cenozoic uplift and climate change in the
stratigraphy of the Macquarie River Valley, New
South Wales. Australian Journal of Earth Sciences
51,273-290.

Truswell, E.M. and Owen, J.A. (1988). Eocene pollen
from Bungonia, New South Wales. In Palynological
and Palaeobotanical studies in honour of Basil E.
Balme (Eds PA. Jell and G. Playford). Association
of Australasian Palaeontologists, Sydney Memoir 5,
259-284.

Tulip, J.R., Taylor, G. and Truswell, E.M. (1982).

Palynology of Tertiary Lake Bunyan, Cooma, New
South Wales. BMR Journal of Australian Geology
and Geophysics 7, 255-268.

Vandenberg, A.H.M. (2010). Paleogene basalts prove early
uplift of Victoria’s Uplands. Australian Journal of
Earth Sciences 57, 291-315.

Van der Beek, P.A., Braun, J. and Lambeck, K. (1999).
Post-Palaeozoic uplift history of southeastern
Australia revisited: results from a process-based
model of landscape evolution. Australian Journal of
Earth Sciences 46, 157-172.

154

Proc. Linn. Soc. N.S.W., 136, 2014

H.A. MARTIN

Veevers, J.J. (1991). Mid Cretaceous tectonic climax, Late
Cretaceous recovery and Cainozoic relaxation in the
Australian Region. In The Cainozoic in Australia: a
re-appraisal of the evidence (Eds M.A. J. Williams, R
De Deckker and A.P. Kershaw). Special Publication
of the Geological Society of Australia 18, 1-14.

Wallace, M.W., Dickson, J.A., Moore, D.H. and Sandiford,
M. (2005). Late Neogene strandlines of southern
Victoria: a unique record of eustasy and tectonics in
southeastern Australia. Australian Journal of Earth
Sciences 52, 279-297 .

Webb, J.A., Finlayson, B.L., Fabel, D. and Ellaway.

M. (1991). The geology of the Buchan Karst
- implications for the landscape history of the
Southeastern Highlands of Australia. In The
Cainozoic in Australia: a re-appraisal of the evidence
(Eds M.A. J. Williams, P. De Deckker and A.P.
Kershaw) Special Publication of the Geological
Society of Australia 18, 210-234.

Wellman, P. and McDougall, I. (1974). Potassium-argon
ages in the Cainozoic rocks of New South Wales.
Journal of the Geological Society of Australia 21,
247-274.

Williams, R.M., Merrick, N.P and Ross, J. B. (1989).
Natural and induced recharge in the lower Namoi
Valley, New South Wales. In M. L. Sharma (Ed.)
Groundwater Recharge pp 238-253 (A.A. Balkema,
Rotterdam).

Williamson, W.H. (1986). Investigations of groundwater
resources of the Lachlan Valley alluvium. Part
l.Cowra to Jemalong Weir. Water Resource
Commission of NSW Hydrological Report 1986/12.

Wooley, D.R. (1978). Cainozoic sedimentation in
the Murray Drainage Basin, New South Wales.
Proceedings of the Royal Society of Victoria, 90, 61-
65.

Young, R.W. and McDougall, I. (1985). The age, extent
and geomorphological significance of the Sassafrass
basalt, southeastern New South Wales. Australian
Journal of Earth Sciences 32, 323-33 1 .

Young, R.W., Young, A.R.M., Price, D.M. and Wray,

R.A.L. (2002). Geomorphology of the Namoi alluvial
plain, northwestern New South Wales. Australian
Journal of Earth Sciences 49, 509-523.

Proc. Linn. Soc. N.S.W., 136, 2014

155

156

Integrating History and Ecological Thinking: Royal National

Park in Historical Perspective

Daniel Lunney

Office of Environment and Heritage NSW, PO Box 1967, Hurstville NSW 2220, and School of Biological
Sciences, University of Sydney, NSW 2006. Email address: dan.lunney@environment.nsw.gov.au.

Published on 27 June 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Lunney, D. (2014). Integrating history and ecological thinking: Royal National Park in historical
perspective. Proceedings of the Linnean Society of New South Wales 136 , 157-199.

This paper aims to develop an ecological history of Royal National Park. The socio-cultural context for
the push to reserve such a large tract of land in perpetuity in 1 879 includes the Park’s early links to the Royal
Zoological Society of NSW (formerly the Acclimatisation Society of NSW), in addition to a strong political
movement advocating the reservation of open space in urban areas. A selection of maps of the Park situates
it in a broader context. Previously unpublished data from 1879 to the present is evidence of increasing
formal support for nature conservation and protected areas. Tim Flannery’s contentious essay ‘Beautiful
Lies’ (2003) is challenged on the issue of long-term fauna conservation in Australia’s national parks. The
paper concludes that using an ecological approach to interpreting historical data enables us to gain a clearer
grasp of the reasons behind the changes to the Park’s boundaries since 1879, the relationship between the
Park and its fauna, and the challenges facing the Park as an urban park in the twenty -hrst century.

Manuscript received 22 August 2013, accepted for publication 23 April 2014.

KEYWORDS: acclimatisation movement, ecological history, environmental history, Royal National Park,
Royal Zoological Society of NSW, Tim Flannery; urban park, Yellowstone National Park.

INTRODUCTION

Oliver Rackham is a scholar with an unusual
bent. He is interested in ancient woodlands and has
developed the area of woodland ecology as a branch
of historical ecology, which he sees as both a science
and a part of history (Rackham 2003: xviii). He points
out that woodland ecology is a discipline that is still
in its early stages of development. In the second
edition of his striking book, Ancient Woodland ',
Rackham notes that new data have strengthened his
conviction that ancient woods are all different, and
that each has its own unique development. Given that
Rackham (2003: 435) views Australia as a miniature
planet and contends that its ecosystems work on
different principles to the rest of the globe, one can
quickly appreciate that, from a world perspective,
Royal National Park is an international treasure richly
deserving of its own ecological history. The Linnean
Society symposium of 2011 was a major step toward
achieving that goal, by examining the Park from a
number of different interpretive positions (see e.g.
Adam 2012; Attenbrow 2012; Schulz and Magarey

2012). This paper aims to further that endeavour by
moving between history and ecology to arrive at a
deeper understanding of the future challenges facing
the park.

Ecological history is a rapidly growing held
attractingconsiderableinternationalattention.Drawing
on existing fields such as environmental history
(with which it is often synonymous) and historical
geography, ecological history has been recognised as
crucial to developing ecological restoration programs
and conservation strategies (Foster 2000; Donlan and
Martin 2004; Jackson and Hobbs 2009), in addition
to deepening our understanding of the human impact
on the natural environment (Flannery 1994). As
an approach, ecological history seeks to integrate
disparate disciplines, drawing not only from ecology
and history, but also cultural studies (Goodall 2010;
D’Arcy 2006) and archaeology (Hayashida 2005;
Briggs et al. 2006), among other fields. Many works
in the held adopt a grand-scale approach, examining
ecological changes which have taken place over
millennia in whole regions (e.g. Vermeij 1 987 ; Flannery
2001; Grove and Rackham 2001). For more localised

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

studies, however, a small-scale approach is equally
valuable in capturing the ecological specificities and
changes in a given area. Although recognising that
the history of Royal National Park - both ecological
and cultural - did not begin with its dedication in
1879, this paper focuses on the decades following its
dedication, which have been underexamined in the
context of ecological history. In her captivating book,
The Colony , historian Grace Karskens identifies that
by the 1 820s, the pattern of fanning and grazing lands
in New South Wales followed the funnel shape of the
plain’s arable soils. As a result, the rough sandstone
country that encircled the plain was avoided. These
once-shunned areas, she remarks, became Sydney’s
four treasured National Parks: Royal, Blue Mountains,
Ku-ring- gai and Sydney Harbour. In Karskens’
view, “their ecologies became the default ‘Sydney
ecologies’” (2009: 21). The landscapes of the arable
soils, such as on the Cumberland Plain, and the rich
alluvial flats, met a different fate: they are Sydney’s
“lost landscapes” (2009: 21). As Karskens recognises,
the chance survival of a handful of areas has come
to retrospectively structure our understanding of
Sydney’s pre-settlement natural environment as a
whole. That Karskens highlights what has been lost
suggests an awareness of the fact that what we have
left, and the knowledge that can be gleaned from it, is
necessarily incomplete.

There are many fascinating aspects of the
ecological history of Royal National Park. Among
these aspects is the meaning of ‘national park’ and
what it meant in Australia in 1879 when what is now
known as Royal National Park came into existence.
Another is the place of the Park in coastal NSW and
the Sydney region from a biologist’s perspective.
What is its vegetation, its fauna, and how do we
manage this national park ecologically? A third area
of interest is the location of the Park in relation to its
immediate surroundings, and the implications of its
location for the management of this larger unit of land.
As an urban park, it is particularly important in the
context of building public support for conservation
initiatives. Developing a pro-conservation consensus
among urban populations is a key challenge facing
conservation organisations more generally, and
promises to reward protected areas if achieved (Trzyna
2003). The location of the Park also poses specific
challenges for its managers. As Conner (2003) argues,
public awareness of the benefits of protected areas is
particularly important with regard to urban parks. As
such, he contends, managers need to promote their
parks’ natural and cultural heritage values and provide
information to potential beneficiaries with a view to
developing broader support for conservation among

urban constituencies (Conner 2003).

While National Parks are always about the
present, they are also about a sense of the past and
the future. Without an examination of their history we
cannot fully comprehend their development; without
an eye on their future, they will not survive. For
those who lack a sense of history, national parks and
protected areas are an impediment to growth, wasted
land which should be converted into something more
useful. This view is manifest in so many areas of
debate, whether concerning the river red gums on the
Murray, the southeast woodchipped forests, or grazing
in the high country, that we should never rest on the
assumption that we have pennanently made the case
for a national parks system that meets all the ecological
criteria that one can find, including how the parks and
reserve system will fare in an era of climate change.
The shining example of Royal National Park helps
sustain that case. We might rest comfortably with the
assumption that no-one will turn Royal National Park
into a new set of suburbs, but we are far from sure that
the remaining remnants of Sydney’s pre-European
vegetation will not be cleared for some development
dream, a growth centre, infrastructure project or just
incremental expansion of existing suburbs. That is
their likely fate, but it ought not to be. To help project
an image of a future Sydney that keeps as much of its
biological heritage as possible, we should continue to
point to Royal National Park. In 1879, it was a great
idea, by 1979, at its centenary, it was a brilliant idea,
and by 2079, it will be seen as a solid gold investment.
Indeed, as the Trustees concluded in their Official
Guide to the National Park of New South Wales , “It is
Time, and Time alone, that will prove the vast value
of this magnificent dowry to the people of New South
Wales” (Elwell 1893:64).

We can now turn to some of the details of
Royal National Park that might capture the attention
of a future ecological historian who has the time to
follow up any ideas and convert a tentative paper
to a solid piece of scholarship. I might add that it is
essential to publish such efforts: I know of too much
material that is unpublished, and that is a tragedy for
those with more than just a passing interest in Royal
National Park, or indeed any other element of our
natural environment. The importance of research and
education concerning the natural history of Royal
National Park become apparent when listening to
people who have spent much of their lives studying
and working in and around the Park. By 2079, these
experts will have died, and as an important part of
the Park’s history it is necessary that we record this
community’s contribution while they are still active
(see Appendix 1 ). A central theme of this paper is to

158

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

draw attention to the need to record the history of all
our National Parks and Nature Reserves, and to place
their history in an ecological context. It is a difficult
and time-consuming task: it took some years for a
group of us to record Nadgee Nature Reserve (Lunney
et al. 2012), but these efforts will be invaluable in the
coming decades.

‘THE LUNGS OF THE CITY’ : A BRIEF HISTORY

The decision to reserve such a large tract of land
merely 25 kilometres from the Sydney CBD must
be contextualised within the increasing concerns
for public health which preoccupied many of the
educated elite of the nineteenth century. For an
intellectual and political milieu that prized public
hygiene, racial purity and vitality, Sydney’s rapid
population expansion presented critical problems
for the future. The city’s sanitation, overcrowding
and pollution attracted growing criticism in the late
1870s, as a State Government enquiry into Sydney’s
health [1885-1877] blamed a high child mortality
rate on inadequate procedures for sewage disposal.
It was as a direct consequence of these concerns that
urban reformer John Lucas addressed the Legislative
Assembly on 19 February 1879:

“The health of the people should be one of the
first objects of all good Governments, and to insure a
healthy, and consequently a vigorous and intelligent
community, it is necessary that all cities, towns,
villages, and such other centres of populations, should
possess parks and pleasure grounds as places of public
recreation.” (Anon., 1879a: 3)

Lucas proposed that a tract of land should be
dedicated exclusively for the purpose of public
recreation - literal “breathing room” - in all of
Sydney’s densely populated suburbs. In their reportage
of Lucas’ address, the Sydney Morning Herald clearly
agreed. While noting that Sydney already had the
Domain, “some small reserves” - such as Moore
Park, dedicated in 1 866 - and “a most noble harbour”,
it contended that these were insufficient: “With all
those facilities for health we had a puny race of young
people growing up in our midst”. Lucas was especially
preoccupied by the long-term effects of overcrowding
and pollution on children, who lacked “sufficient fresh
air to give them a healthy and vigorous constitution.”
As a result, he viewed the probable consequences of
population expansion “with horror”. In his view, the
Herald reported, “unless provision were made for
sanitary improvements, ... the death rate would be
ten times as much as it was in Sydney at the present
time” (1879a: 3).

Despite the reservations of then Premier Henry
Parkes, who concurred with the sentiment of Lucas’
address but criticised its radical implications for land
use policy, Lucas’ resolution was unanimously passed
in the Assembly the following month. His proposal
sheds valuable light on one of the ways in which
the natural environment was conceived at the time:
as ‘the city’s lungs’, the antithesis to the polluted
urban centre of the modern age. Yet the reformers’
preoccupation with population health was not the sole
factor behind the dedication of Royal National Park
in 1879. As Pettigrew and Lyons (1979) argue, one
of the primary reasons for its reservation from sale
was the need to provide land for the acclimatisation
of foreign animals. The Parkes Government strongly
approved of the aims of the Zoological Society of
New South Wales (initially called the Acclimatisation
Society), which formed a month after Lucas’ address
on 24 March 1879. The Society was committed to
“the introduction and naturalisation of song-birds,
and of animals suitable for game” (Anon. 1879b: 5).
Two days after its first meeting, the Sydney Morning
Herald reported that the Parkes Government, “in order
to promote its objects, will set apart a large tract of
land for the purpose of acclimatisation.” It specified
that “the proposed reserve is on the south side of Port
Hacking, extending from the coast some five miles
back, and is said to embrace about 80,000 acres”
(Anon. 1 879c: 5). On 29 March, the Herald described
the area in greater detail and credited the idea to John
Robertson, Vice-President of the Executive Council,
“who has thought of the project for years” (Anon.
1879d: 3). 18,000 acres (7284ha) were formally
dedicated on 26 April 1879. On the same day, eleven
Trustees were appointed, including Lucas, Robertson,
and the convenor of the Zoological Society, Walter
Bradley.

That Lucas and Bradley were both appointed as
Trustees points, to a certain extent, to the compatibility
of their aims. Both men, and the groups they
represented, viewed the natural environment within
a utilitarian framework. Although today the effective
cooperation of a zoological body and an urban
development group is complicated by the former’s
conservation ethic, in the nineteenth century their
objectives were far more complementary. Irrespective
of their individual backgrounds as naturalists,
urban refonners, and government officials, the first
generation of Trustees shared an understanding of the
National Park as a reserve which existed primarily for
public use. Its central purpose was to provide a space
for public recreation. Accordingly, the Trustees saw
the ‘beautification’ and ‘improvement’ of the Park as
high on their list of management priorities. Central

Proc. Linn. Soc. N.S.W., 136, 2014

159

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

to this was the key problem of accessibility. Over the
Park’s first decade, the Trustees devoted the majority
of their funding and effort toward the provision of
access routes (Fig. 1). For some tenderers, clearing
areas of the Park proved too great a challenge. As
one tenderer, John Crowley, writes to his contractor
in 1882: “I beg to inform you that I am reluctantly
compelled to decline proceeding with the clearing
portion of the National Park [. . . ] I am not surprised at
having been deceived in my estimate of the work as the
undergrowth of gum and appletree [. . . ] are all suckers
growing from stumps of saplings and large trees that
have been burnt level with the natural surface of the
ground” (State Records NSW, Container No: 9/2188).
Despite such setbacks, the 1893 Guide boasts that,
during the Trust’s first five years, “thirty-two miles
of roads were cleared, and a considerable length was
fonned and finished for traffic.” With the growing
popularity of the Park as a “recreation resort”, road
construction operations were extended. “From that

day to this,” the Guide continues, “the work of road
formation has been continued, and in the main,
satisfactorily completed”. The result is a network
of “thoroughfares, now spreading web-like over the
park” (Elwell 1893: 12-13).

These operations were applauded by the
public. Although part of the Park’s allure was that
it had “remained so long unknown, unvisited, and
unappropriated” - indeed, a “terra incognita” - it was
considered inevitable that it would be “subdued to
the hands of man” (Anon. 1879e: 4). As the Sydney
Morning Herald commented: “In the main it is as little
known and has been as little visited as if it had been
1000 miles away. The time has come for this solitude
to be disturbed.” In the reporter’s estimation, this was
“simply the rescue from neglect of a beautiful piece of
wild country, and bringing it forth for the enjoyment
of man” (Anon. 1879e: 4). Tellingly, the enthusiastic
public response to the decision to reserve the Park in
March 1879 was strongly linked to the expectation

Fig. 1. Audley Road, National Park (Government Printing Office, 1888). Photograph courtesy of the
National Library of Australia (Digital Collection; Call Number ‘PIC/8476/13 LOC Album 1037’).

160

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

that it would accelerate the planning for the long-
awaited Illawarra railway line. With this means of
transportation, the park would be “a sanctuary for
the pale-faced Sydneyites, fleeing the pollution,
physical, mental and social, of that densely packed
city.” [quoted in Pettigrew & Lyons (1979) but no
source cited. M. Maack (2002) attributes quote to
John Robertson; “NSW Confederation Conservation
History”, The Bushwalker, Vol. 28, No. 1 (August
2002), p. 3],

The 1887 Deed of Grant formalised the Trustees’
responsibility to the public. It empowered the Trustees
“to use and permit to be used the said lands as aNational
Park for the recreation of the inhabitants of the said
colony” and specified the Park’s legitimate uses.
These included “ornamental plantations of lawns and
gardens”, “zoological gardens”, an “artillery range”
and the “exercise and encampment of military or naval
forces” (N.S.W. 1891: 3). The rest and recreation of
the public were high on the list of priorities (Fig. 2).
In alignment with the broader utilitarian philosophy

which underscored the management of the Park, the
Deed clarifies that the Park’s natural resources are
subservient to public need. It continues:

“. . . it shall be lawful for the Trustees of the National
Park to grant licenses to mine upon and under the
said land for and to take away and dispose of, as the
licensees may think fit, all coal, lime, stone, clay,
brick, earth or other mineral (excepting gold or silver)
that may be found in the said lands.” (1891 : 4)

In her work-in-progress, entitled European
history of Royal National Park revisited ', Judith
Carrick examines the history of attempts to mine the
Park in more depth than can be explored here (Carrick,
in press: 18-20). For our purposes, it is illuminating
to note that the dominant conception of the Park as
a space for public use coexisted in relative harmony
with a deep appreciation of its perceived beauty.
There seems to have been little concern, for example,
when the Park’s tableland was extensively cleared in

Fig. 2. Unknown boy on banks of river, National Park (Charles Bayliss, ca. 1880-1900). Photograph
courtesy of the National Library of Australia (Digital Collection; Call Number ‘PIC/7985/164 LOC
Album 100’).

Proc. Linn. Soc. N.S.W., 136, 2014

161

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Fig. 3. Encampment Ground, Loftus Heights (Government Printing Office, 1888). Photograph cour-
tesy of the National Library of Australia (Digital Collection; Call Number ‘PIC/8476/11 LOC Album
1037’).

1884 for use by the military (Fig. 3). That this was
considered a routine matter of management and not
environmental degradation points to the historical
specificity of the naturalists’ relationship to the fauna
and flora which they studied with fervour. However,
despite these disturbances the Park retains many of its
biodiversity values (Adam 2012; Schulz & Ransom
20 1 0; King 2013) and continues to meet contemporary
criteria for designation as a National Park.

POINTS OF CONVERGENCE: THE
ZOOLOGICAL SOCIETY OF NSW

To a twenty-first century ecologist, the attitudes
and priorities of the nineteenth-century naturalists
seem bizarre. Particularly incomprehensible is the
Zoological Society’s interest in the acclimatisation
of foreign species at a time when native fauna and

flora had not yet received comprehensive legislative
protection. Indeed, the first statute enacted in NSW
addressing the issue of fauna protection, the Animals
Protection Act of 1879, listed as its primary purpose
the “importation and breeding” of alien species. The
protection of native fauna (the list of which includes
no mammals) rated second - and only applied “during
the breeding season” (N.S.W. 1879: 56).

This stipulation reflects the acclimatisation
movement’s selective approach to the issue of
preservation more generally. It supported the protection
of certain native fauna on the basis of its utility. As
Pettigrew and Lyons (1979: 18) argue, its proponents
believed that the contemporary rates of exploitation
had to be regulated not for conservation purposes, but
to ensure that there remained sufficient populations for
future generations to exploit. Furthermore, although
it was largely comprised of naturalists passionate
about the natural environment, the acclimatisation

162

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

movement was, from the vantage-point of the
contemporary conservationist, quite arrogant: it
believed it could ‘improve’ nature. Few saw anything
problematic in this objective; on the contrary, many
were drawn to it by a sense of boundless possibility
As the Sydney Morning Herald commented, “It is
difficult to set limits to the attractiveness which this
fortunate national reserve may be made to possess”
(Anon. 1879e: 4).

This conception of the Park, and of the fauna and
flora within its shifting boundaries, persisted well into
the twentieth century. The Official Guide of 1914, for
example, is redolent with references to ‘beautification’,
and boasts of the successful introduction of multiple
non-indigenous species, including trout and perch.
Yet the Guide hints at an introduction which would
prove a headache: that of deer at Gundamaian. By
1886, the Trustees had acquired seven fallow deer,
some white angora goats, and five valuable red deer

through donation. According to the 1 893 Guide , they
thrived and rapidly multiplied (Elwell 1893: 13). A
special Deer Park was established to house the deer
near the Port Hacking River (Fig. 4), and more deer
were later purchased. Yet, as the Guide records, “for
them nine- wired fences did not a prison make. [...]
these ruminants broke bounds, and are now roaming,
fancy free, over the wide domain” (Elwell 1893: 54).
As early as 1 893, Carrick notes, there was a complaint
about deer escaping and destroying a neighbouring
garden. By 1912, the Trust refused an offer of more
deer, and by 1923 the Trust was attempting to ‘donate’
them to other parks. The management of deer,
particularly the Javan rusa, remains a most difficult
issue to this day (Keith and Pellow2005).

Gundamaian was also home to the Scientists’
Cabin. According to Carrick, the Cabin was built in
1924 for the Zoological Society, although Allen Keast
remembers that it “had formerly housed the timber

Fig. 4. Fountain cottage and the fountain at the Deer Park, Port Hacking River (Government Print-
ing Office, 1888). Photograph courtesy of the National Library of Australia (Digital Collection; Call
Number ‘PIC/8476/4 LOC Album 1037’).

Proc. Linn. Soc. N.S.W., 136, 2014

163

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

workers” engaged in logging operations before the
Society occupied it in the “late 1920s” (Keast 1995a:
28). It was in the vicinity of the sawmill by the Hacking
River, just above the Upper Causeway. During its time
there, the Society conducted valuable research into the
native birds of the Park, particularly the bower-bird.
The Society was granted sole use of the cabin, but
was given notice to leave in 1935 because it could not
agree to the new terms of the permissive occupancy.
Carrick notes that records show that the Society was
still there in 1941. Concerning the eviction of the
Zoological Society, Keast bitterly recalls that “the end
of the Cabin came ignominiously about 1944 when
most members of the Society were absent at the war:
it was pilfered bit-by-bit for seaside cottages on the
adjacent Park beaches” (1995a: 29).

Another interesting point of convergence between
the Park and the Zoological Society is the push within
both for the addition of the prefix ‘Royal’ to their
titles. As public recognition of their value grew, so did
their stature. To the management of both the Park and
the Society, the insertion of ‘Royal’ would suitably
reflect their growing importance in the eyes of the
public. By 1908, almost 30 years after its fonnation,
the Society had risen in prominence to the extent
that its President, Dr. T.P Anderson- Stuart, sought
permission to add the ‘Royal’ prefix to the name. A
Royal Charter was duly granted in September 1908.
On 10 February 1909, the Society changed its name to
‘The Royal Zoological Society of New South Wales’.
Three decades later, the Trustees of the National Park
discussed renaming it to Royal National Park, while
other parks (namely, Ku-ring-gai) would be National
Parks. For the Park, it was the visit of Queen Elizabeth
II in 1954 which would prove to be decisive: the Park
was renamed in 1955.

The addition of ‘Royal’ can be interpreted as both
a political and cultural statement. It is distinctively
British, it carries certain class overtones, and it was
a fashion statement which the Royal Easter Show,
Royal Society for the Prevention of Cruelty to Animals
(RSPCA) and the Royal Flying Doctor Service also
reflect. Its connotations raise the question of exactly
which sectors of colonial society were to benefit from
the dedication of the National Park in 1 879. It is indeed
far from clear as to whether the National Park was
dedicated for the poor of the inner suburbs for health
and recreation, or for a more privileged group that
could consider importing and releasing exotic species,
the very ones we now call alien invasive species.
One should read the press release and accompanying
documents with a critical eye. For historians, there
is some digging to do here, particularly concerning
the meaning of ‘national’. Carrick, for example,

argues that the word ‘national’ denoted, in 1848, the
inclusion of all individuals in a locality irrespective
of denomination and social standing. In view of this,
one could reasonably extrapolate that, in 1879, the
National Park was dedicated for all inhabitants of the
colony. At the very least, it was certainly understood
this way: the Sydney Morning Herald ', for example,
makes few references to Sydneysiders when speaking
of the Park’s use, preferring inclusive language such
as “the people of the whole colony” and “the people of
this country” (Anon. 1879d: 3; Anon. 1879e: 4).

On a related note, it is difficult to discern whether
John Robertson was inspired by a foreign model when
he created Royal National Park - and if so, which one.
There is certainly some merit to the claim that the
isolated Yellowstone would be an odd model for a park
located so close to the inner city (Pettigrew and Lyons
1979). It is more likely that if Robertson had a model,
it was London’s recently established common parks,
located on the border of the metropolis, though this is
in need of further research. In Carrick’s view, the links
between the American trajectory and developments in
Australia are ambiguous and in need of more probing
study. In my perspective, the debate over which
national park was first in the world - Yellowstone
in 1872, or our local candidate - is distracting. It is
more productive to examine the claims to originality
critically and within their political context, as Robin
(2012) has done in an intelligent paper. The ecological
ideas of the 1870s (not, of course, conceived in
twenty-first century ecological terms), are equally
interesting. Their echoes are still present in NSW,
whether the current public debate centres on marine
parks, mining under the parks (such as the coal seam
gas proposals for the Pilliga forests in north-western
NSW), or hunting on public lands.

MAPPING ROYAL NATIONAL PARK

We live in a tenure-bound society. Maps are a
manifestation of our preoccupation with boundaries,
and of our specific relationship to the natural
environment, although they have a long history of use
in navigation. They are today so commonplace that it is
difficult to grasp their initial novelty. The early maps of
Royal National Park were among the first in Australia
of a natural area enclosed by a boundary for the sake
of demarcating an area considered to be purely natural.
Until these maps were designed, natural history in
Australia did not have set boundaries within which the
natural environment could be managed. The mapping
of Royal National Park fundamentally challenged the
dominant exploitative approach to the land as a place

164

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

to be colonised, cleared, and farmed. It gave emphasis
to an emergent perspective of the natural environment
which was not primarily valued in commercial terms
and which was beginning to recognise, by the late
nineteenth century, that forests could not be exploited
in an unregulated manner (Lunney & Moon 2012).
That this was a public area owned and managed by the
State in perpetuity remains one of the great landmarks
in world nature conservation. Royal National Park
initiated the integration of nature study with the
management of natural areas. In so doing, it made an
extensive part of New South Wales’ pre- settlement
environment accessible to a large number of people
who otherwise may not have come into contact with
some of the most beautiful specimens of Australian
fauna and flora in their natural settings.

For these reasons, it is worth turning our
attention to the maps of Royal National Park. Many
observations can be gleaned from examining the
maps in sequence and in context of the surrounding
areas. In what follows, I examine a series of maps
chronologically in order to draw out some of the
factors which have contributed to the dedication of
the Park, and to illustrate the changes in the Park’s
boundaries and management over time.

The earliest map of the area of relevance to this
study is dated 1 845 , and depicts the “country southward
of Sydney, shewing the Road lately opened through it
to the Illawarra” (Fig. 5). Operationally speaking, this
road came to define the Park’s boundaries, prefiguring
the western border of the Park. Thatthere were no roads
in this area prior to 1 845 can be seen as evidence that
the land was of little commercial value: in comparison
to the arable soil of the Cumberland Plain, for example,
the land which was later to comprise Royal National
Park had not been opened up for grazing crops or
farming estates. Consistent with this, the map shows
a clear absence of landscape differentiation, with no
references to ownership. Indeed, it resembles more
an explorer’s map than the careful result of a set of
surveyor’s decisions. As Surveyor-General, Thomas
Mitchell (whose signature appears at the bottom of
the map) would have well understood the importance
of tenure boundaries as a reflection of political and
administrative decisions regarding land use. The
absence of tenure boundaries on this map points to the
fact that, in 1845, there had been no decisions made
on the potential use of this area of undifferentiated
Australian bush. Instead, it had escaped 57 years of
colonisation without being surveyed and considered
for agricultural and commercial use. With Sydney
growing in a pattern that fitted the arable lands, it was
a chance of geography, soil fertility, and the ready
access to more productive landscapes that allowed the

future Park area to remain ‘unused’ (in a contemporary
land use sense) until 1 879. Consistent with this, in the
earliest existing parish map of Wattamolla - undated,
but appraised to have been constructed between 1835
and 1870 - the sentence “barren land destitute of
timber” was inscribed across what we now know as
Royal National Park (Fig. 6). This phrase remained in
subsequent maps lithographed in the early 1 870s (Figs.
7-8). This indicates that, for successive governments,
this land had been surveyed and had no commercial
value.

In contrast to the 1 845 map, a map of the Park
dated 1879 (Fig. 9) displays clear tenure boundaries
in the typical block fashion, with the leaseholders’
names printed neatly on their respective portions (a
close-up of this part of the map is provided in Fig. 10).
This map was found by Allan Fox “crumpled in the
corner of a room” in Royal National Park in the late
1 970s, when Fox was helping to assemble information
for the celebration of the Park’s centenary. Fox states
he found it “in a pile of rubbish to be thrown out”
(pers. comm, 2013). The land which it depicts is
representative of the original boundary which became
the area dedicated in 1879 in three jiarts: the first on
the 26 m of April, the second on the 6 tn of October, and
the third on the 25 th of November (NSW Government
Gazette 1879b, 1879c, 1879d). The upper left-hand
corner of the map states that it is a “tracing shewing
National Park &c., County of Cumberland”. Although
the map does not provide the name of the surveyor, it
resembles an official document, perhaps prepared in
readiness for the Park’s dedication that year. Given
the pencil marks on the map, it has the appearance
of a working map. Interestingly, the words “Reserve
from sale pending selection of railway line” cover
a large area on the Park’s western boundary which
would later be excised (see below). Most importantly,
however, the tenure boundaries of this map provide us
with a timeframe within which to assess the changing
management of this area of land. They show us that,
between 1845 and 1879, decisions were being made
on its potential use. What in 1 845 had no formal land
use designation was beginning to be dissected in 1 879
for other uses. In view of this, it becomes clear that
had the decision not been made to dedicate the area as
a National Park, the vast majority of this land would
have been cut up into private holdings by the turn of
the century.

The gazettal notice of 26 April 1879 states that
18,000 acres were dedicated and gives a detailed
written description of the boundary (N SW Government
Gazette 1879b). We have used contemporary GIS
technology to draw the boundary according to this
original description. The calculated area stands at

Proc. Linn. Soc. N.S.W., 136, 2014

165

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Fig. 5. ‘Country southward of Sydney, shewing the road lately opened through it to the Illawarra’.
Sydney: Thomas Mitchell, 1845. Map reproduced courtesy of the Mitchell Library, State Library of
New South Wales. Call number ‘Cb 84/18’.

166

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Fig. 6. Wattamolla parish map. Circa 1835-1870. Map reproduced with permission of the NSW Lands
and Property Information, Department of Finance and Services, Panorama Ave., Bathurst 2795.

19,541 acres (7908 ha) and forms the basis of a new
map, shown in Fig. 1 1 . According to the gazettal notice
of 3 August 1880, the Park was expanded on this date
by 19,000 acres (NSW Government Gazette 1880).
Again following the gazettal description, we used GIS
technology to calculate the total area to be 36,532
acres (14,784 ha) and the actual area is depicted in
Fig. 11. This largely - though, as we will see, not
completely - fonns the basis of what is now known
as Royal National Park. The addition in 1880 is in the
eastern half and incorporates the land which is shown
in Fig. 10. It appears that the allotments shown in Fig.
1 0 were mining leases (as indicated by the initials ‘ML’
in the corner of each portion), leading us to assume

that either the terms of the lease had lapsed by 1880,
or that the approval for mining had been withdrawn.
The absorption of these allotments may thus shed
light on the early Trustees’ relationship to mining in
the Park: as Mosley (2012:35) has suggested, John
Robertson and his supporters may have gone to great
lengths to protect the Park from this threat.

For our purposes, it is interesting to note that the
Park was still being surveyed at this time at Robertson’s
request (State Records NSW, Container No: 9/2188).
In June 1879 a representative of the Department
of Lands, PT Adams, opined that “on survey
considerable modification of the present boundaries
will be found necessary”, and argued that “natural

Proc. Linn. Soc. N.S.W., 136, 2014

167

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

iftpTacuridte-

R&trm - / ’"/./

from ■ t >cr/c d/uf Tun fee
hi censr*

. \i;f iff eel 0 /, fj,e

Hariaar .«* A/'rif I-S 7 S ■

iifuif eif jffeaeh

Fig. 7. ‘Parish of Wattamolla in the County of Cumberland’. Circa 1873-1874. Map repro-
duced with permission of the NSW Lands and Property Information, Department of Finance
and Services.

FA HISH

Wi TTAMQLU

■ rl.<:

‘CjfAfUti

RES

: ( •»' f
JlF ff 4*1- JcZtc iWtL'AiJ >L-'

fyrffrvn rsr.fttn.- Xxasjhd- i'F

, acdrf-'/i'rt Jtbritiitf s v .fWs, ;i a . v-.*v t

\,Vb iitjed' ' '

Ttecich

'Hes$rvF

LiUftilteS A

'■■■$ g-i

\ijUFM -Skst/if &0>f

Mi.

mo. j /
»raa»? C

gigs

Fig. 8. ‘Parish of
Wattamolla in the
County of Cum-
berland’. Circa
1880-1882. Map
reproduced with
permission of the
NSW Lands and
Property Infor-
mation, Depart-
ment of Finance
and Services,
Panorama Ave.,
Bathurst 2795.

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

she*

PARK &c

CUMBERLAND

TH WALES

Sorts

Souct

jfou-a

fochf'Jbmt

THbtf

Sbmtrwns P‘

' LPdicatiPfi

Mfoleotfarn

. P^ <r — j -w

rlChLAN

W M

A?3 J?/

P/8 as

^71 -'5

.TWclt

fc \r Jfolt ppdta*

^va^IKY K

/u.i. . IhtfT 5

wfifA&r p i:
j •K«z.d/ :

|/.W*

JfoOtty

mbit L
m /oj 7,

.1/ /

• Jd*<icr
\Y J' J/Wa.,

Tko*£o(uifM

Ofwtm

T.-i', lutur IfrlVM *» J
'* "71

7* r *rttiul*n trr^ftrf7l

W. > <*“■

j&w& / t

Fig. 9. ‘Tracing shewing National Park &c., County of Cumberland, New South Wales, 1879.’ Un-
published map. Reproduced courtesy of Allan Fox.

Proc. Linn. Soc. N.S.W., 136, 2014

169

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Fig. 10. Tenure boundaries (detail), 1879. Image taken from Fig. 9. Unpublished map. Reproduced
courtesy of Allan Fox.

170

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Legend

26 April 1879
6 October 1879
25 November 1879
3 August 1880

Sketch: Guide. Map

Shewing Rivers Creeks .Roads, Bridle Paths &c.

Port Hacking River

County of

New South Wales

J anuahv 1861

Stale about J-Miie a I Inch.

Memoranda

dddsf &r j

yZwty, *$*&£. «,/ M$

vJtar/^j/m 'MlS es

di-nA -'ph-ttys m

' /d £

List of Roads opened

/ fa-sdsdf -ds/ardsf*/' m jj-sd/ ,/s,

Ajds. trm., '_jur.^ n - , vrfL, i. ra' 1 ^.r« A hS , A?r'ur**£,

//•rKsrfa-' ,-

y.M &,jj

id kttJ&d,*.

.. (Lird-'id- d-Z/umstl.-'d- /^IWw 'tavCiW®' A*A

Vj'&mfir** &id> ’ , iLd/,/ /%/- ofSdL^a/

lii-Hx/ fe4' >£ jtU#u/-^ ^.iid/tay.//fddt^.dnr»9t' 'xs.dr: t'JLt/,; oaua,.- A

*•***£& tfjLH'J?; J&m/.A Jbdni $Zw ^

r &tdfdi/*rM

■4 J d

pSJL*.d . ^' .

J jk™ £ am y ? d •sdi^.dYmxdvwj e^dt- ’-/id/

’adiy/itf ,i*-d.-'t^d‘?'.

J-JdLs %Ll

-jCsr.<j-' Asdr' 'Ti'-rdry.-!/.-

- iM-Ytita- dim- Y/:". 7 dxi/s A- yjjy itnjA-j- f . /

., tdditq/ ^ ? drtdj.A jfbukJji . . _

*»ii M-»(l, dad- ijjdmis'lj d/ J&jJ-

LO CAL A KET CH

JffElWffC? T#£ /?Ci5 TO W* Jfax'fltoli P>173>T

P«

Fig.

11. Map showing incremental additions to Royal National Park from 26 April 1879 - 3 August 1880. This
map uses the 1881 map in Fig. 12 as a base to show the boundaries of the successive increments over
this period. The land bounded by the red line is the initial dedication of 26 April 1879. The green line
shows the addition of 6 October 1879. The yellow line shows the addition of 25 November 1879. The land
within the boundary of the Park which falls outside these lines was gazetted on 3 August 1880. This map
was constructed using GIS to map the written descriptions in the gazettal notices of the four dates listed
above. This approach allowed an accurate determination of the total area of each increment.

Proc. Linn. Soc. N.S.W., 136, 2014

171

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

features should be substituted when the [sic] exist
for arbitrary lines” (State Records NSW, Container
No: 9/2188). In January 1881, a “sketch guide map
shewing rivers, creeks, roads, bridle paths &c” in the
newly dedicated National Park was lithographed at
the Surveyor General’s Office in Sydney (Fig. 12).
When placed alongside the 1879 map, this map clearly
illustrates the expanded boundary of the Park. In the
1 881 map, the Park has absorbed the private holdings
depicted in Fig. 7. Moreover, the information at the
bottom of the map states that the “area of the park is
approximately 36,000 acres” (Fig. 12). This is double
the figure given in the Government Gazette for the
Park’s size in 1879, noted as “18,000 acres” (NSW
Government Gazette, 1879a, 1879b). With a series
of acquisitions (NSW Government Gazette 1880),
the Park’s southern boundary now roughly followed
a line between Garie Beach and what later became
Waterfall. Given that there existed no precedent for
determining the boundaries of National Parks, it is
understandable that the area doubled so early on, as
competing uses of the land may have been resolved in
the early years of the Park’s administration. However,
it is remarkable that this considerable expansion has
largely gone unnoticed in the existing histories of the
Park. These early changes to the Park’s boundaries
are worthy of a separate study, as deeper examination
of how and why they occurred may shed light on the
colonial administration’s understanding of the Park
in its earliest years. For our immediate purposes,
however, it suffices to note that the fluidity of the
Park’s boundary in its early decades reflects the
fluidity of the concept of a ‘National Park’ at this
juncture. While we repeatedly cite 1 879 as the pivotal
year of dedication, it is in actuality only the first stage
in the Park’s history, and is representative not of a final
boundary, but of an initial area set to greatly expand.

An official map dated 1897 (Fig. 13) provides
us with another point of departure in examining the
developmental history of the Park. Interestingly,
this map appears to be identical to a map dated 1 893
and published as part of the Official Guide (Elwell,
1893). The map clearly depicts the location of the
Illawarra railway line in the area which was marked
‘reserved from sale’ in Fig. 9. Furthermore, the area
west of the railway line is shown to remain within
the Park’s boundary. This was not to last long,
however: as the politician and editor Andrew Garran
presciently noted in 1886, “though it may remain a
wild preserve, the railway will soon bring the long
line of southern suburbs close up to its edge” (Garran
1974 [1886]: 98). The NSW Government Gazette of
26 August 1903 confirmed his prediction, declaring
the intentions of the Governor, “with the advice of the

Executive Council of [NSW]”, to “wholly revoke the
said dedications and grant in so far as they apply to
or affect the said areas of 36 acres, 54 acres, 5 acres,
13 acres and 2 roods, 2 acres and 2 roods, and 2,950
acres of land described in the Schedule hereto” (NSW
Government Gazette 1903: 6293-6294). A total of
3,060 acres was excised from the Park’s western
boundary. The Park’s new boundary is shown in an
official map produced in 1904 (Fig. 14). Interestingly,
it appears that this map was a personal copy owned
by the architect and conservationist Myles J. Dunphy,
who was later to become known for his tireless efforts
to protect key areas of the Blue Mountains. The 1904
map states that the area of the Park is now 33,719
acres - down from 36,320 in 1897. According to
Carrick, the Park’s Trustees agreed to a proposal made
in 1 895 by the Lands Department to withdraw this
area, and received Jibbon Reserve (shown in Fig. 13
to be excluded from the Park) in exchange (Carrick, in
press: 7). This is consistent with Carrick’s contention
that a “symbiotic relationship” existed between the
Trustees and the Department of Railways “from the
beginning” (Carrick, in press: 42), and is worthy of
further research in a future study.

These maps illustrate considerable changes
to the Park’s boundaries in its early decades. Yet,
although these changes are directly observable when
represented visually, they are often discussed in the
aggregate in existing literature. This has confused our
understanding of the historical development of the
Park. For the purposes of clarifying this development,
a number of graphs and tables were prepared for this
paper. Cathy Johnson of the Reserve Establishment
and Land Information Section (OEH) prepared
a spreadsheet tracking the 37 additions to Royal
National Park over the period 1 October 1967 - 11
March 2005, increasing the park size from 14,851.94
ha on 1 October 1 967 to its current size of 1 5,09 1 .7 1 73
ha (Table 1 ). Appendix 2 provides a crucial context
for appreciating the information provided in Table 1 .
It shows the date of dedication, initial area, and area
modifications of all of the National Parks and Nature
Reserves in NSW prior to the National Parks and
Wildlife Act 1967. While accessible, the information
provided in this appendix is extremely difficult to
locate and, to the author’s knowledge, has not been
reproduced. It is appended here as a benefit to scholars.
While providing area information in two or more
decimal places may seem too fastidious, precision is
vital in view of the vulnerability of Australian parks and
reserves more generally. There is the issue, however,
of whether the surveys were sufficiently accurate to
justify this many decimal places. As we have now
established that, in 1879, the figure of 18,000 acres

172

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Fig. 12. ‘Sketch Guide Map shewing Rivers, Creeks, Roads, Bridle Paths, &c. National Park. Port Hack-
ing River, County of Cumberland, New South Wales.’ Sydney, New South Wales: Surveyor General’s
Office, January 1881. Map reproduced courtesy of the Mitchell Library, State Library of New South
Wales. Call number ‘Z/Cb 88/3’.

Proc. Linn. Soc. N.S.W., 136, 2014

173

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Fig. 13. ‘Plan of the National Park: shewing Railway Stations, Roads, &c.’ Sydney, New South Wales:
Department of Lands, 1897. Map reproduced courtesy of the Mitchell Library, State Library of New
South Wales. Call number ‘Z/Ml 811.114/1897/1’.

was simply a close approximation, what is at issue
now are the incremental additions and revocations
to Royal National Park, as shown in Table 1 and Fig.
15. We can reasonably assume that, since 1967, any
further changes were mapped with a higher level of
accuracy and thereby provide interested parties with
clear and precise infonnation. Furthermore, in view of
these standards, this paper adopts the current reporting
level of accuracy.

From an historical viewpoint, the records do not
begin at OEH, or the National Parks and Wildlife
Service [NPWS], before October 1967 when the

National Parks and Wildlife Act 1967 was passed
and the NPWS established. The National Parks and
Wildlife Act 1974 replaced the earlier Act, and is
the current Act under which Royal, and indeed all
the National Parks and Nature Reserves in NSW,
are acquired and managed. Mike Prentice (also of
the Reserve Establishment and Land Information
Section) and Cathy Johnson kindly helped me to
isolate the specific additions to and excisions from
the Park. Their data were used to construct a series of
maps, which illustrate the changing boundary of the
Park from 1879-201 1 (Figs 16a-h). As these maps are

174

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

COUNTY OF CUMBERLAND

MEW S©U™ WALES

1904

AREA 33716 ACRES

ESTATE

SUTHERLAND

SOUTH ftjp,

audLey:

MAfU-f.y RF.‘

/ SKETCH

SHEWING RELATIVE POSITION OF

SYDNEY AND NATIONAL PARK

BULOO,M-

HELENSBURGH

EXPL.AKA TORY NOTES ■

4 : ]j ff?

Cw ” y

r-T / '■

\ JfZ;

] y |

...

SUTHERLAND

:5k

~=ry

i x:

JLzz:

r“ m

\'* = i

iS> - •

MAP OF THE

Cxi

-T

NATIONAL PAEK

Fig. 14. ‘Map of the National Park, County of Cumberland, New South Wales’. Sydney, New South
Wales: Department of Lands, 1904. Map reproduced courtesy of the Mitchell Library, State Library of
New South Wales. Call number ‘Z/M3 811.114/1904/2’

Proc. Linn. Soc. N.S.W., 136, 2014

175

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Table 1. Additions to Royal National Park, 1 Octo-
ber 1967 - 11 March 2005. Credit: Cathy Johnson
(Reserve Establishment and Land Information
Section, OEH).

Legend

Date

Area (ha)

01 -Oct- 1967

14,851.94*

A-l

06-Dec- 1968

19.24

13-Dec- 1968

~1H9

06-Jun-1969

HIT

05-Dec- 1969

~U19

05-Dec- 1969

TT5T

A- 7

08-May- 1970

8.77

08-May- 1970

“TTT4

A-9

24-Dec- 1970

4.9852

24-Dec- 1970

“TJT5

A- 11

28-Jan-1972

A98

A-l 2

04-Feb-1972

“ATS

13-Oct-1972

“TITS

A-l 4

28-Sep-1973

“TT9

“AT5

07-Dec- 1973

T73

31 -May- 1974

UT5

A-17

29-Nov-1974

“5T2

“ATS

19-Mar- 1976

TOO

“AT9

08-Oct-1976

“ATS

“AT0

17-Nov-1978

TOO

A-22

28-Dec- 1979

TTTO

“AT3

23-Oct-1981

~7M

A-24

21 -Jan- 1983

TT3

18-Feb-1983

0.26

23-Dec- 1983

“039

06-Jun-1986

TITS

ll-Sep-1987

"TTT7

“A30

26-Feb-1988

TT3

“ATT

18-Mar- 1988

03-Feb-1989

15-Dec- 1989

“4TU0

“ATT

25-Jul-1997

“TTTf

01 -Dec-2000

042

A-36

1 1 -Mar-2005

TB1

consistent in scale and visual presentation, they are
provided as a supplement to the original maps, which
can be difficult to compare.

Drawing upon all available data for all Parks
and Nature Reserves in NSW, a detailed graphical
presentation of the growth of the Parks and Reserves
system from its inception in 1879 to the present (30
June 2012) is shown in Fig. 17. These graphs place
the maps of Royal National Park in the historical

context of growing support for the dedication of
Parks and Reserves in NSW. The standard way of
displaying the Parks and Reserves in NSW is in map
form, meaning that unless one compares one map to
another, the growth pattern is not easily discernible.
Furthermore, unless one digs through the Gazette
records, their growth in area is not apparent, especially
when the growth in a given period is comprised of a
series of modest increments. The regular mapping of
the distribution of Parks and Reserves in NSW has
missed the value of the pattern of numerical growth
over time. These graphs were prepared specifically
to overcome this deficiency in existing scholarship,
and represent a new contribution to our understanding
of the Parks and Reserves system in NSW. The data
which were used to construct these graphs is provided
in Appendix 3.

There are many points that can be made from the
documents provided in this section. By examining the
sequence and the dates, it is remarkable that such a
large area was explicitly named as a National Park in
1879. There had been many small parks set aside in and
around Sydney for recreation and health, but nothing
near the size of Royal National Park at that time. It
then becomes surprising that the area doubled in size
so quickly: indeed, it has grown, with the additions of
the adjacent Heathcote and Garawarra expanding the
conservation estate. The growth of the Park in recent
decades parallels the growth of the National Parks
estate across NSW. Interpreting the earliest additions
to the Park’s boundary is complex, however: it is
possible that they reflect growing support for the
Park, and lobbying by special interest groups such as
the Acclimatisation Society.

When a series of other maps are set next to
these documents, further features emerge that help
explain why the Park area remained Crown land in
1879. The maps in Benson and Howell (1990: 8) and
Keast (1995b) depict the area as sandstone plateau
country unsuitable for fanning. Thus, it was a chance
of geography that left the area intact for the 91 years
since the European settlement of Sydney. However,
it would not have been there in 1967 when the first
National Parks and Wildlife Act came into force. Early
timing was thus crucial in its dedication. The issue of
timing, the importance of Crown lands, and the role
of government attitudes have all been recognised as
factors which determine the acquisition of national
parks and nature reserves. This recognition largely
grew out of the Fauna Protection Panel in the 1950
and 1960s and, after 1967, the NSW National Parks
and Wildlife Service. Exactly which of these factors
took precedence in the acquisition of National Parks
in twentieth-century NSW was the subject of a debate

176

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Legend

^ Additions Greater than 1 Hectare Road

NPWS Reserve Waterbody

Royal National Park
Additions Greater than 1 Hectare

Prepared by RET
27 February 2014

rtjjtil Office of

jQjyy Environments Heritage

South Pacific Ocean

F - Otford & Stanwell Tops

Towra Point NR

E - Lilyvale

Port Hacking

Bate

Bay

A - Kirrawee & Grays Point

South Pacific Ocean

C - Bundeena

D - Heathcote

B - Lilli Pilli

Port Hacking

Fig. 15. Additions to Royal National Park greater than one hectare, 1967-2005. This map was produced
by Cathy Johnson (Reserve Establishment and Land Information Section, OEH) specifically for this
paper. For the data to which it refers, see Table 1.

in 1990 in the A iistralicm Zoologist (McMichael 1990;
Pressey 1990; Reed 1990; Specht 1990; Starling 1990;
Whitehouse 1990a,b). For so many areas, the decades
between World War II and the turn of the century
held the last chance to dedicate large new parks and
reserves, and it remains one of the feats of foresight
and action that we have such a magnificent set of
national parks and nature reserves in NSW.

FAUNA AND HISTORY

In the Official Guide to the National Park of New
South Wales (Elwell 1 893), the Trustees also comment
on the history of Royal National Park in relation
to its fauna. The Trustees credit themselves with
making great strides in fauna conservation, as a direct
consequence of preventative by-laws which prohibit

Figs 16a-h (next 4 pages). The changing outline of Royal National Park from 1879-2011. Comparing the
maps allows us to discern a number of changes to the Park’s boundaries over time: most significantly, a
large excision from the Park in its north-west corner (indicated by colouring the excised land grey), and
a steady expansion of the Park’s area. The names of key places are provided for reference purposes. Fig.
16a shows the outline at 26 April 1879. Fig. 16b shows the outline on 6 October 1879, after an addition
to the Park. Fig. 16c shows the Park’s boundary in 1881. Fig. 16d shows the Park’s boundary in 1904,
with the grey area representing an excision from the Park’s area since 1881. Fig. 16e shows the outline in
1967, when Royal became part of the NSW National Parks and Wildlife Service. It also shows the area
of Heathcote National Park, which was dedicated in 1963. Fig. 16f shows the additions to Royal National
Park between 1968-1979. Fig. 16g shows the additions to Royal between 1980-1999 and the location of
both Heathcote National Park and Garrawarra State Conservation Area. Fig. 16h shows the current (at
30 June 2012) boundary of Royal National Park, which had been established by 2000. These maps were
produced by Cathy Johnson (Reserve Establishment and Land Information Section, OEH) specifically
for this paper.

Proc. Linn. Soc. N.S.W., 136, 2014

177

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

16 a above, 16 b below

6 October 1879
25 November 1879
ROYAL NATIONAL PARK

Mariey Beach

Helensburgh

1.25 2.5

Kilometres

Legend

Stanwell Park

6 October 1879
25 November 1879
n Royal National Park

178

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

16 c above, 16 d below

1904

0 1.25 2.5 5

Kilometres

Otford

Stanwell ParK

Legend

n Royal National Park

Proc. Linn. Soc. N.S.W., 136, 2014

179

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

16 e above, 16 f below

-•C.ronulla

Wattamolla

•Helensburgh

■Otlord

1968 - 1979

ROYAL NATIONAL PARK
HEATHCOTE NATIONAL PARK

Marley Beach

Garie

Legend

□ Reserve Addition
| | Royal National Park

n Heathcote National Park

180

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

•Cronulla

■Engadine

■Bundeena

Wattamolla

•Stairwell Park

1980 -1999

ROYAL NATIONAL PARK
HEATHCOTE NATIONAL PARK

v__ — J

GARAWARRA STATE
CONSERVATION AREA

Marley Beach

Garie

Legend

Reserve Addition
Royal National Park
Garawarra State Conservation Area
Heathcote National Park

;Zgn

16 g above, 16 h below

2000 - PRESENT

ROYAL NATIONAL PARK

HEATHCOTE NATIONAL PAI^K

GARAWARRA STATE
CONSERVATION AREA

Marley Beach

Legend

Reserve Addition

□ Royal National Park

□ Garawarra State Conservation Area
n Heathcote National Park

Proc. Linn. Soc. N.S.W., 136, 2014

181

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Fig. 17 (next 3 pages). The growth of NSW National Parks and Nature Reserves from the dedication of
Royal National Park in 1879 to the end of the 2011-12 financial year.

Fig. 17 has been produced as a series of graphs, numbered 1-5. Graph 1 provides a context for the
four subsequent graphs, which have been constructed using the same data but are produced on different
scales corresponding to the area involved. Graph 2 shows the area of National Parks from 1879 until the
formation of the NSW National Parks and Wildlife Service in 1967, which integrated the selection and
management of National Parks and Nature Reserves into one organisation. Note that the scale has been
adjusted from Graph 1 to show more detail, in accordance with the smaller areas of National Parks prior
to 1967. Graph 3 shows the area of Nature Reserves from 1955-1967, dedicated under the Fauna Protec-
tion Act 1948. Graph 4 shows Graphs 2-3 combined. Graph 5 shows the growth of both National Parks
and Nature Reserves in the period 1967-2012. In Graphs 2-5, one vertical axis shows area and the other
shows the percentage of NSW that is dedicated as National Parks and Nature Reserves. These graphs
are original. While the information to construct the 1897-1967 graphs is formally available, it is difficult
to locate. However, we were able to construct these graphs due to the expert help of Mike Prentice and
Cathy Johnson of the Reserve Establishment and Land Information Section (OEH), where meticulous
records are kept. The details of the dates and area of dedication of National Parks and Nature Reserves
prior to 1967 is given in Appendix 2. These are presented here to provide easy access to these data.

Graph 1: National Parks Estate in NSW

Generated on March 08, 2013 at 1 :34:05 PM

182

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Graph 2: Area of National Parks up to 1967

Year

Generated on March 08, 2013 at 1 :34:05 PM

Graph 3: Area of Nature Reserves up to 1967

year

Set up on March 08, 2013 at 1 :34:06 PM

Proc. Linn. Soc. N.S.W., 136, 2014

183

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Graph 4: National Parks plus Nature Reserves up to 1967

Year

Generated on March 08, 2013 at 1 :34:06 PM

Graph 5: National Parks plus Nature Reserves since 1967

Generated on March 08, 2013 at 1 :34:06 PM

184

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

“the exposure of articles for sale” and the hunting of
both native and introduced fauna (Elwell 1893: 17).
The responsibility for enforcing these policies lies
with “all employees of the Park Trust”, who have been
“sworn in as special constables” and are henceforth
“enjoined and empowered” to ensure their effective
implementation. In the Trustees’ estimation, they
have been successful:

“This policy of preservation is already achieving the
desired results, for the National Park is now the haunt
of a great variety of beautiful birds. [...] The almost
extinct lyre-bird, free from molestation, can be daily
seen, about sunrise and sunset, seeking its food among
the brush glades and stately ferns on the banks of Bola
Creek. Now and again the satin-bird, the regent-bird,
the rifle-bird, all famed for their beauty of plumage,
and which, in their wild state, are becoming rarer and
rarer owing to the insatiable and wanton cruelty of
prowling hoodlums and men of higher degree who
degrade the name of sportsman, can be seen flitting
from tree to tree in some of the deeper recesses of this
guarded reserve.”

The paragraph concludes with the only mention
of mammals: “A few marsupials remain. Sometimes
on a still night the eerie howl of the dingo can be heard
on the lonely mountain sides, and the handsomely-
marked native cat has been known to leave evidences
of nocturnal depredations” (Elwell 1893: 17-18).

A number of points about fauna management
can be drawn from these notes. The most striking
is that, by 1914, the National Park was seen as a
sanctuary for animals. This was not mentioned in
1879. This development could be taken to reflect the
influence of the Zoological Society and its interest
in acclimatisation, and the particular interests of its
convenor and amateur ornithologist Walter Bradley
- one of the original Trustees. Either fauna was an
unheralded initial interest in setting up National
Park, or it was a concern that did not come to fruition
until shortly after the park was established and
professionally managed. The next point of interest
is that the Trustees recognised the incompatibility
of stock and national park aims. However, the loss
of fauna beyond the park was laid at the door of the
hunter, not the clearing of land, nor the running of
cattle and sheep. What is evident is the pride in the
fact that the National Park did hold birds of such
beauty that the Trustees knew would gain public
approval. The phrase ‘almost extinct’ shows an
insight into what fate a species might face if not
protected. Although it is unclear whether this phrase
refers to the state of the lyre-bird population within

the Park or within Australia more widely, the use of
such language is remarkable given that, at the time, no
working knowledge existed of the extinction of any
Australian vertebrate. Although Gould recognised
that the numbers of certain species were declining,
and recognised the possibility of total disappearance,
there remains no evidence of any knowledge of past
extinction. There were few laws that protected fauna:
despite broadening the scope of protection offered
to specific fauna, the Birds and Animals Protection
Act 1918 was in many ways ineffective (Stubbs
2001), and it was not until the Fauna Protection
Act 1948 that native birds and mammals received
widespread legislative protection which provided
for the establishment of faunal reserves. Jarman and
Brock (2004) provide a history of these laws and the
evolution of the concept of ‘endangered species’ .

As an ecologist with a particular interest in fauna
conservation, I look at Royal National Park in a
regional context, with a particular interest in the koala
Phascolarctos cinereus. Royal National Park does not
hold the high quality habitat that koala populations
need to survive. It does hold patches of koala habitat,
but it is the land in and around Campbelltown, to the
west of Royal National Park, with an arc of land to
the south, that carries koala habitat, and indeed a
koala population that has been there continuously
since European settlement (Lunney et al. 20 10a, b).
Koalas can literally walk from Campbelltown to
Royal National Park; indeed, tagged koalas have
demonstrated this ability. In view of this, the Park
can be recolonised, with the major barrier being the
Princes Highway, a killing zone on the western edge
of Royal National Park. It is the position of koalas in
Royal National Park, or the current lack of koalas, that
Tim Flannery has targeted to expose what he sees as
the weakness of our national parks in regard to wildlife
conservation. Flannery’s argument is brief: “If we look
around at our national parks today, what we see in the
great majority of cases are marsupial ghost-towns,
which preserve only a tiny fraction of the fauna that
was there in abundance two centuries ago. A classic
example is Royal National Park south of Sydney. It’s
the nation’s oldest park, yet over the last few decades
it has lost its kangaroos, its koalas, its platypus and
greater gliders. Clearly, it is a fallacy to believe that
proclaiming more such reserves will do very much to
preserve Australian wildlife.” (2003:39)

My interpretation of koala distribution is that
it is much more tied to factors such as soil fertility,
watered lands and nutrient-rich leaves. The lands
which fit these criteria are now mostly agricultural
lands, which have largely been cleared so that habitat
loss is the primary cause of the decline of the koala

Proc. Linn. Soc. N.S.W., 136, 2014

185

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

in NSW (Reed et al. 1990; Reed and Lunney 1990).
Koala conservation is an issue for land use planning
to protect koala habitat on private land, as is the
management of other threats, such as fire, dog attack,
disease and death on the roads, as stated in the NSW
2008 Koala Recovery Plan (DECC 2008). There are
plans in place to tackle these matters and the Senate
(2011) recently released its findings into the health
and status of the koala. It is concerned for its future in
Australia. It did not, however, identify the supposed
failure of national parks as a problem. Flannery is
right to point out that we cannot rely solely on national
parks and nature reserves to conserve all our wildlife,
for their conservation does depend heavily on the
lands with the rich soils, which of course were the
ones cleared early and so comprehensively (Lunney
& Moon 2012). However, it is hard to read such a
sensible cautionary note into Flannery’s sentences. It
is easier to read his text as being dismissive of parks
and reserves for conserving wildlife. In this regard, his
argument is reminiscent of the criticism of the reserve
system contained in the Commonwealth’s 1996 State
of the Environment report, which contended that as the
existing system did not reflect terrestrial biodiversity,
it had “only limited value as an antidote” to the
threats facing biodiversity (Commonwealth 1996: 49;
Lunney 1998).

What is alarming is the logic of the leap that
Flannery makes from saying that Royal National
Park had lost its koalas to arguing that proclaiming
more reserves will not do very much to preserve our
wildlife. Flannery had not established that koalas
were there at first settlement, or ever flourished there.
Partly, this is due to the fact that the fauna records of
the Park are patchy and heavily weighted towards
recent decades (although he does not acknowledge
this). This evidentiary deficiency, however, does
not of itself justify Flannery’s conclusions. I have
yet to find an early record, but my general thesis is
that koalas were not likely to be present so close to
the coast because a large population of Aboriginal
people, mostly living on the food from the sea and
the estuaries, would have hunted any local koalas to
extinction. Locations further from the coast, such as
Campbelltown, or the adjacent locations, Bargo and
Nattai, where the koala was first seen in Australia by
Europeans, are more likely because, in my conjecture,
the local Aboriginal population would have been at
a lower density. The appearance of koalas in Royal
National Park may well reflect the loss of the local
Aboriginal population of hunters.

It does seem to be a limited argument to select
a few large mammals, consider them to be extinct in
one location, and thereby write off all the parks and

reserves for wildlife conservation. We might note too
that Royal National Park was not set up on modern
ecological principles for wildlife conservation. Why
write off all the national parks and nature reserves
on the basis that the first national park in Australia
does not hold all of its original fauna? By all means,
Flannery can point to the limitations of our parks
and reserves for wildlife conservation so that we
continue to tackle all the issues facing our fauna,
but those limitations present, in my view, no case
for abandoning what I regard as the best means we
have ever devised for fauna conservation. There is
no surprise that the NSW environment minister Bob
Debus should reply to Flannery and state: “Let me
rebut Dr Flannery’s plainly ridiculous allegation that
the Royal National Park. . . is a ‘marsupial ghost town’ .
[. . . ] On the contrary, the NPW S is able to demonstrate
that the Royal National Park does in fact provide
important habitat for numerous small marsupials.” He
added, “In any event, Royal National Park does not
exist in isolation. It is on the very edge of a continuous
reserve system that runs for hundreds of kilometres”
(2003:114).

The kerfuffle over Flannery’s paper raises a
number of important points. It shows that we do
need to examine the history of an area to be able to
interpret it ecologically. Arguably Flannery blundered
with koalas because he knew too little about the
history of Royal National Park, the specific context
for its dedication in 1879, and the history of koala
management in Australia. In 1879, koalas, along with
other native fauna, were shot for the fur trade and as
pests. Lunney and Leary (1988) document the koala
fur trade at the end of the 1 9th century for the Bega
district in the Eden region of NSW, and Gordon and
Hrdina (2004) document the millions of koalas shot for
the fur trade in Queensland in the early 20th century.
Given these research findings, it would seem odd to
propose that the species was in need of reservation of
land. As the early accounts of the Park reveal, it was
the Park’s beautiful birds and plants that first captured
the imagination. Ecological history does rely on
getting the historical part of the equation right before
one can speculate successfully on the cause and effect
of change in wildlife numbers and distribution. The
koala story of Royal National Park has not yet run its
course, but it will, in my view, not support Flannery’s
thesis.

Further evidence which challenges Flannery’s
thesis has been provided by a number of koala
sightings in and around the area of Royal National
Park. Park rangers have reported finding a deceased
koala, initially released at Kentlyn on the west side of
the Georges River on 29 July 2012. By late September

186

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

2012, the koala had returned to the Sutherland area
and was found dead on the western side of the Princess
Highway. Additionally, a local resident living in
Kirrawee photographed a koala in September 2012
(Fig. 18) from the balcony of their house, located on
the northern boundary of the Park adjacent to Savilles
Creek. According to Park employee Glenn Harvey
(pers. comm 2013), the koala has been observed in
this area for “the past couple of years”. Furthermore,
she states that the koala has also been sighted in the
Kirrawee High School grounds and “slightly further
north on Hunter Street”. Harvey also reports recent
sightings of “two koalas at Deer Pool” on 23 March
2013 and of one koala crossing the road at McKell
Avenue (near the Park toll box) on 25 March 2013.
She states that these were “credible” but unconfirmed
sightings.

These sightings demonstrate that koalas inhabit
the area to the Park’s west and are within walking
distance of the Park. This koala population is a
continuous population that inhabits Campbelltown
and tagged koalas have been recorded as walking
as far as Campbelltown to the western edge of the
Park (Lunney et al. 2010a). The fact that koalas
occur within the Park but have not proliferated is
evidence that Royal National Park is essentially not
koala habitat. Thus, one could conclude that the Park
has not Tost’ its koalas but that, instead, it never had
them in abundance. A similar story is emerging for
the greater glider (Petaroides volans). Andrew et al.
(in press) detail the reappearance of the glider after
its presumed disappearance in recent decades. Royal
National Park was never known for containing many
greater gliders, and the extensive fires of 1994 may
have eliminated the small population from the Park.
This work points to the fact that this glider species was
never a common animal in the Park, but is capable of
reaching the Park. Thus the Park has, once again, not
lost its greater gliders, for it did not (excepting small
patches) provide high-quality glider habitat in the first
place. More broadly, this points to the importance of
conserving lands which encompass the full range of
habitats in a state, including the fertile lands which
support species such as the koala and the greater
glider.

Whatever Flannery’s views on the parks and
reserves system may have been in 2003, he declares
strong support for it in his latest essay ‘After the
Future’. He contends that “the creation of the national
parks system must surely be seen as the principal
environmental achievement of the past half- century”
(2013: 26). His comments show that even those who
criticise the parks and reserves system on the basis of
the ‘CAR’criteria-comprehensiveness, adequacy, and

Fig. 18. A koala sighted at Gore Avenue, Kirrawee,
New South Wales (19 September 2012). Photo-
graph by Erin Meagher.

representativeness - still recognise the intrinsic value
and significance of the system. Given the frequency
of such criticisms, even among ecologists, it is crucial
to acknowledge that the parks and reserves system
is an evolving idea. Consequently, one’s judgement
regarding its adequacy needs to be tempered by an
historical perspective which recognises the importance
of context. Our focus should lie on how the system
might be improved, rather than on its shortfalls in
view of contemporary ecological criteria.

FUTURE THREATS

At the ‘Transforming Australia’ conference in
July 2011, Flannery launched the report on the climate
change forecast for the NSW south coast (Climate
Commission 2011a). This impact statement was
accompanied by ‘The Critical Decade’ report (Climate
Commission 2011b). Both proj ect an array of worrying
impacts. Alongside concerns for biodiversity and the
increasing vulnerability of coastal towns due to rising
sea levels, the report notes that higher temperatures
will increase the likelihood of large and intense fires
in the region. At particular risk are areas such as the
Royal National Park and the forested escarpment
behind Wollongong, including the Woronora Plateau.

As Mooney, Radford and Hancock (2001)
demonstrate, fire has long been an issue for the Park,
with significant fire events occurring throughout the
twentieth century. This raises the issue of scale. In our
study of the impact of the 1994 bushfires on the koala
population at Port Stephens, we concluded that koalas
rapidly re-occupy the burnt forest within months, and
are breeding in the forest by the next breeding season
(Lunney et al. 2004, 2007). The issue was not how
many hectares were burnt, but how many were left.

Proc. Linn. Soc. N.S.W., 136, 2014

187

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

In the case of Port Stephens, the fire consumed only
half of the koala habitat, so recolonisation was rapid,
with individual koalas walking up to 1 km per day.
For Royal National Park and its non-flying fauna, the
central concerns are where the refuges lie, how to
manage them, and the fire history of these sites. Fire
history is an integral part of an ecological history of
an area. Movement from nearby areas is possible, but
the barriers, particularly the major roads, are an issue
of considerable significance. For koalas, re-colonising
from Campbelltown is possible, but greatly hindered
by the barrier of the Princes Highway. In this context,
the park can be seen as too small for some species, but
not a ‘ghost-town’.

The International Union for Conservation of
Nature’s selection of Australia as the host of the
2014 World Parks Congress reflects a growing
international recognition of the global significance
of Australia’s parks. It is therefore opportune to place
Royal National Park in historical and ecological
perspective. Given that the Park has existed for 135
of the 226 years since the European settlement of
Australia, it reflects enormous changes in Australian
society; indeed, it can be taken as a barometer of
social and political attitudes, especially in regard
to the development of a conservation ethic. In the
years since its dedication, our understanding of what
constitutes a national park has undergone a distinctive
intellectual shift. This has paralleled a transformation
of our understanding of fauna conservation and land
use, and the role of government in the management
of land. It is tempting to examine Royal National
Park solely from an historical perspective or an
ecological one; what is more novel is integrating the
two interpretive frameworks in order to understand
what the dedication of the park signified in 1 879, and
how this has since changed. For the Park, this has
meant analysing a variety of sources, including maps,
records of fauna, media reports, and statistical data.
Looking at the environment of the park in the context
of its socio-political history, as a major part of our
first steps toward nature conservation, and in view of
future threats, all point to the necessity of integrating
historical and ecological thinking.

ACKNOWLEDGEMENTS

1 am indebted to many colleagues over the 43 years 1
have worked for OEH and its predecessors, particularly the
NSW National Parks and Wildlife Service for their insights
and appreciation of Royal National Park. In particular, I
thank Mike Prentice, Cathy Johnson, Murray Robinson and
Rob Dick for their information regarding the dates of park
dedication and area size, and to biometrician Ian Shannon

for the excellent graphs of the growth of national parks
and nature reserves in NSW since 1879. 1 also thank Cathy
Johnson for her efforts in creating Fig. 11 and Figs. 16a-g.
I am indebted to Judith Carrick for kindly allowing me to
draw on her soon-to-be-published history of Royal National
Park. Finally, 1 am indebted to Antares Wells for applying
her skills as an historian to reshaping this manuscript and
thus enabling the more effective integration of historical and
ecological strands of thinking.

REFERENCES

Adam, P. (2012). Royal National Park - Lessons for the
Future from the Past. Proceedings of the Linnean
Society of NSW 134 , 7-24.

Andrew, D., Koffel, D., Harvey, G., Griffiths, K., and
Fleming, M. (In press). Rediscovery of the Greater
Glider Petauroides volans (Marsupialia: Petauroidea)
in the Royal National Park, NSW. Australian
Zoologist 36, X-X.

Anon. 1879a. Places of Public Recreation. The Sydney
Morning Herald (19 February): 3. Anon. 1879b. News
of the Day. The Sydney Morning Herald (25 March):
5.

Anon. 1879c. News of the Day. The Sydney Morning
Herald (26 March): 5.

Anon. 1879d. ANational Park. The Sydney Morning
Herald (29 March): 3.

Anon. 1879e. Untitled. The Sydney Morning Herald (2
April): 4.

Attenbrow, V. (2012). The Aboriginal Prehistory and

Archaeology of Royal National Park and Environs: A
Review. Proceedings of the Linnean Society of NSW
134 , 39-64.

Benson, D., and Howell, J. (1990). ‘Taken for granted:

The bushland of Sydney and its suburbs’. (Kenthurst,
N.S.W. : Kangaroo Press).

Briggs, J., Spielmann, K., Schaafsma, H., Kintigh, K.,
Kruse, M., Morehouse, K., and Schollmeyer, K.
(2006). Why ecology needs archaeologists and
archaeology needs ecologists. Frontiers in Ecology
and the Environment 4 , 1 80-188.

Carrick, J.A. (2014). ‘History of Royal National Park’.
(Published by Judith Carick, Austinmeer, NSW,

Austral ia)(page numbers in text refer to the pre-
publication draft).

Climate Commission Secretariat [Department of Climate
Change and Energy Efficiency], (2011a). The Critical
Decade: Illawarra / NSW South Coast Impacts
[Online], Available at: <http://climatecommission.
gov.au/wp-content/uploads/4246-CC-Wollongong-Key-
Messages_web.pdf> [Accessed 04/03/20 13],

Climate Commission Secretariat [Department of Climate
Change and Energy Efficiency], (2011b). The Critical
Decade: Climate Science, Risks and Responses
[Online], Available at: <http://climatecommission.
gov.au/wp-content/uploads/The-Critical-Decade_
July36revision_Low- res.pdf> [Accessed 04/03/2013],

188

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Commonwealth of Australia. (1996). ‘State of the

Environment. Australia. 1996’. (Collingwood, VIC
: CS1RO). Available at: <http://www. environment,
gov. au/soe/ 1 996/publ ications/report/index.html>
[Accessed 10/04/2013],

Conner, N. (2003). Some benefits of protected areas for
urban communities: A view from Sydney, Australia.

In ‘The Urban Imperative: urban outreach strategies
for protected area agencies : how those responsible for
protected areas can better serve people in large cities
and build stronger urban constituencies for nature
conservation’ (ed. Ted Trzyna) pp. 34-43 (Sacramento
: InterEnvironment, California Institute of Public
Affairs, 2005).

D’Arcy, P. (2006). ‘The People of the Sea: Environment,
Identity, and History in Oceania’ (Honolulu:
University of Hawai’ i Press).

Debus, B. (2003). Beautiful Lies: Correspondence. The
Quarterly Essay 11, 112-116.

Department of Environment and Climate Change

NSW. (2008). ‘NSW 2008 Koala Recovery Plan’.
Department of Environment and Climate Change,
Sydney, N.S.W. Available at: <http://www. environment.
nsw.gov.au/resources/threatenedspecies/08450krp.pdf
[Accessed 14/02/2013],

Donlan, C. J., and Martin, P. S. (2004). Role of Ecological
History in Invasive Species Management and
Conservation. Conservation Biology 18, 267-269.

Elwell, T. (1893). ‘An Official Guide to the National Park
of New South Wales; with map denoting roads, Port
Hacking River, and Port Hacking, creeks, brooks and
interesting localities, specially prepared views of
picturesque scenery, and a general index. Published
by authority of the Trustees’ (Sydney: Charles Potter,
Government Printer).

Farnelf F., Carruthers, J.H., Lee, C.A., McMillan, W.,
McGowen, J.S.T., Wise, B.R., Sullivan, P.H., Ashton,
J., Macfarlane, J., Hogue, J.A., Nobbs, J., Mahony,
W.H., McDonald, R., Nielsen, N.R.W., and Cann,

J.H. (1914). ‘The Official Guide to the National Park
of New South Wales’. (Sydney: Government Printer).

Flannery, T. (1994). ‘The future eaters: an ecological
history of the Australasian lands and people’ .

(Chats wood, N.S.W. : Reed Books).

Flannery, T. (2001). ‘The eternal frontier: an ecological
history of North America and its peoples’.

(Melbourne, Australia: Text Publishing).

Flannery, T. (2003). ‘Beautiful lies: population and

environment in Australia’. (Melbourne, Vic.: Black
Inc.).

Flannery, T. (2013). After the Future: Australia’s New
Extinction Crisis. Quarterly Essay 48, 1-80.

Foster, D. (2000). Conservation lessons and challenges
from ecological history. Forest History Today (Fall),
2 - 11 .

Fox, A. (2013). Pers. comm. [29 May],

Garran, A. (ed.) (1974). ‘Australia, the first hundred years:
being a facsimile of Volumes I & II of the Picturesque
Atlas of Australasia, 1888’. Sydney: Ure Smith.

Goodall, H. (2010). Nets, backyards and the bush: the
clashing cultures of nature on the Georges River. In
‘The Natural History of Sydney’ (eds. D. Lunney,

P. Hutchings, and D. Hochuli) pp. 35-43 (Mosman,
N.S.W. : Royal Zoological Society of New South
Wales).

Grove, A., and Rackham, O. (2001). ‘The Nature of

Mediterranean Europe: An Ecological History’. New
Haven; London: Yale University Press.

Hayashida, F. (2005). Archaeology, ecological history, and
conservation. Annual Review of Anthropology 34,
43-65.

Harvey, G. (2013). Pers. comm [19 April],

Hrdina, E., and Gordon, G. (2004). The Koala and

Possum Trade in Queensland, 1906-1936. Australian
Zoologist 32, 543-585.

Jackson, S. T., and Hobbs, R. J. (2009). Ecological
Restoration in the Light of Ecological History.

Science 325, 567-568.

Jarman, P, and Brock, M. (2004). The evolving intent and
coverage of legislation to protect biodiversity in New
South Wales. In ‘Threatened species legislation: Is
it just an Act?’ (eds. P. Hutchings, D. Lunney, and
C. Dickman) pp. 1-19 (Mosman, N.S.W. : Royal
Zoological Society of New South Wales).

Karskens, G. (2009). ‘The Colony: a history of early
Sydney’. (Crows Nest, N.S.W. : Allen & Unwin).

Keast, A. (1995a). The Sydney ornithological fraternity,
1930s-1950: anecdotes of an admirer. Australian
Zoologist 30, 26-32.

Keast, A. (1995b). Habitat loss and species loss: the birds
of Sydney 50 years ago and now. Australian Zoologist
30,3-25.

Keith, D., and Pellow, B. (2005). Effects of Javan Rusa
Deer ( Cervus timorensis) on Native Plant Species in
the Jibbon-Bundeena Area, Royal National Park, New
South Wales. Proceedings of the Linnean Society of
NSW 126, 99-110.

King, R.J. (ed.) (2013). ‘Field Guide to Royal National
Park, New South Wales’. (Kingsford, N.S.W. :
Linnean Society of New South Wales).

Lunney, D. (1998). The role of national parks and nature
reserves as a field for science. In ‘National Parks: new
visions for a new century’ (ed. Peter Prineas) pp. 140-
149 (Sydney, N.S.W. : Nature Conservation Council
of New South Wales).

Lunney, D. (2010). A history of the debate (1948-2009)
on the commercial harvesting of kangaroos, with
particular reference to New South Wales and the role
of Gordon Grigg. Australian Zoologist 35, 383-430.

Lunney, D., Close, R., Bryant, J., Crowther, M S.,

Shannon, T, Madden, K., and Ward, S. (2010a).
Campbelltown’s koalas: their place in the natural
history of Sydney. In ‘The Natural History of Sydney’
(eds. D. Lunney, P. Hutchings, and D. Hochuli)
pp. 319-325 (Mosman, N.S.W. : Royal Zoological
Society of New South Wales).

Lunney, D., Close, R., Bryant, J., Crowther, M S.,

Shannon, I., Madden, K., and Ward, S. (2010b). The
koalas of Campbelltown, south-western Sydney: does

Proc. Linn. Soc. N.S.W., 136, 2014

189

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

their natural history foretell of an unnatural future? In
‘The Natural History of Sydney’ (eds. D. Lunney. P.
Hutchings, and D. Hochuli) pp. 339-370 (Mosman,
N.S.W. : Royal Zoological Society of New South
Wales).

Lunney, D , Fox, A., Catling, P, Recher, H., and Lunney,

H. (2012). A contribution to the ecological history of
Nadgee Nature Reserve, on the south coast of New
South Wales. In ‘Australia’s Ever- changing Forests:
Proceedings of the eighth National Conference
on Australian Forest History’ (eds.B. J. Stubbs,

J. Lennon, A. Specht, and J. Taylor) pp. 95-124
(Canberra, A.C.T. : Australian Forest History Society).

Lunney, D., Gresser, S.M., Mahon, PS. and Matthews,

A. (2004). Post-fire survival and reproduction
of rehabilitated and unburnt koalas. Biological
Conservation 120 , 567-575.

Lunney, D., Gresser, S., O’Neill, L. E., Matthews, A.,
and Rhodes, J. (2007). The impact of fire and dogs
on koalas at Port Stephens, New South Wales, using
population viability analysis. Pacific Conservation
Biology 13 , 189-201.

Lunney, D., and Leary, T. (1988). The impact on native
mammals of land-use changes and exotic species in
the Bega district (New South Wales) since settlement.
Australian Journal of Ecology 13 , 67-92.

Lunney, D., and Moon, C. (2012). An ecological history of
Australia’s forests and fauna, 1770-2010. In ‘World
Environmental History’ (eds. Mauro Agnoletti,
Elizabeth Johann and Simone Neri Serneri], in
Encyclopedia of Life Support Systems (EOLSS),
developed under the Auspices of the UNESCO
(Eolss Publishers: Oxford, UK). Available at: <
http://greenplanet.eolss.net/EolssLogn/mss/C09/E6- 1 56/E6-
1 56-26/E6- 1 56-26-TXT.aspx# 1 ,_lntroduction_>
[Accessed 10/04/20 13],

Maack, M. (2002). NSW Confederation Conservation
History. The Bushwalker 28 , 3.

McMichael, D. F. (1990). The selection of land for

nature conservation purposes in New South Wales.
Australian Zoologist 26, 78.

Mooney, S., Radford, K., and Hancock, G. (2001). Clues
to the ‘burning question’ : Pre-European fire in the
Sydney coastal region from sedimentary charcoal and
palynology. Ecological Management and Restoration
2 ( 3 ): 203-212.

Mosley, G. (2012). ‘The First National Park: ANatural For
World Heritage’. (Sutherland, N.S.W. : Sutherland
Shire Environment Centre).

N.S.W. Government. (1879). ‘AnActto secure the

protection of certain Birds and Animals’. Available at:
<http://www.legislation.nsw.gov.au/sessionalview/
sessional/act/ 1 879-6a.pdf>[Accessed 14/02/13],

N.S.W. Government. (1891). ‘Deed of Grant, National
Park’. (Sydney, N.S.W.: Charles Potter, Government
Printer.

N.S.W. Government Gazette. (1879a). No. 34. 31 March
1879. 1384.

N.S.W. Government Gazette. (1879b). No. 148. 26 April
1879. 1923-1924.

N.S.W. Government Gazette. (1879c). No. 42. 6 October
1879.4482.

N.S.W. Government Gazette. (1879d). No. 42 extension.

25 November 1879. 5249.

N.S.W. Government Gazette. (1880). No. 314. 3 August
1880.2799.

N.S.W. Government Gazette. (1903). No. 442. 26 August
1903. 6293-6294.

Pettigrew, C., and Lyons, M. (1979). Royal National Park:
a history. In ‘Australia’s 100 Years of National Parks’.
(N.S.W. National Parks and Wildlife Service, Sydney,
N.S.W.).

Pressey, R. L. (1990). Reserve selection in New South
Wales: Where to from here? Australian Zoologist 26 ,
70-75.

Rackham, O. (2003). ‘Ancient woodland: its history,
vegetation and uses in England’. (Dalbeattie:
Castlepoint Press).

Reed, P. (1990). An historical perspective on: Conserving
What? - The basis for nature conservation reserves
inNew South Wales 1967-1988. Australian Zoologist
26 , 85-91.

Reed, P, and Lunney, D. (1990). Habitat loss: the key
problem for the long-term survival of koalas in New
South Wales. In ‘Koala Summit: Managing koalas in
New South Wales’ (eds. D. Lunney, C.A. Urquhart
and P. Reed) pp. 9-31 (Hurstville, N.S.W. : NSW
National Parks and Wildlife Service).

Reed, P, Lunney, D., and Walker, P. (1990). Survey of
the koala Phascolarctos cinereus (Goldfuss) in
New South Wales (1986-87), with an ecological
interpretation of its distribution. In ‘Biology of the
koala’ (eds. A.K. Lee, K.A. Handasyde, and G.D.
Sanson) pp. 55-74 (Chipping Norton, N.S.W. : Surrey
Beatty and Sons).

Robin, L. (2013). Being first: why the Americans needed
it, and why Royal National Park didn’t stand in their
way. Australian Zoologist 36 , 321-329.

Schulz, M., and Magarey, E. (2012). Vertebrate Fauna:
a Survey of Australia’s Oldest National Park and
Adjoining Reserves. Proceedings of the Linnean
Society of NSW 134 ,

Schulz, M., and Ransom, L. (2010). Rapid fauna habitat
assessment of the Sydney metropolitan catchment
area. In ‘The Natural History of Sydney’ (eds. D.
Lunney, P. Hutchings, and D. Hochuli) pp. 371-401
(Mosman, N.S.W. : Royal Zoological Society of New
South Wales).

Senate of the Commonwealth of Australia. (2011).

‘The status, health and sustainability of the koala
population. The koala - saving our national icon’.
(Canberra, A. C.T., Australia).

Specht, R. L. (1990). Comment on Conserving What?

- The basis for nature conservation reserves in New
South Wales 1967-1988. Australian Zoologist 26 ,
76-77.

Starling, J. (1990). Comments on an article by J.
Whitehouse. Australian Zoologist 26, 79-80.

190

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

State Records NSW. (Undated). Miscellaneous papers re:
proposed national park. Series: 10723. Item: 9/2188
(1 box, undated).

Stanley, H. (cl976). History of Royal National Park.
Unpublished manuscript. Available at Office of
Environment & Heritage Library, Hurstville NSW
2220 .

Stubbs, B. (2001). From ‘Useless Brutes’ to National
Treasures: A Century of Evolving Attitudes towards
Native Fauna inNew South Wales, 1860s to 1960s.
Environment and History 7, 23-56.

Trzyna, T. (ed.) (2003). ‘The Urban Imperative: urban
outreach strategies for protected area agencies: how
those responsible for protected areas can better
serve people in large cities and build stronger urban
constituencies for nature conservation’ . Proceedings
of a workshop at the Fifth World Parks Congress,
Durban, South Africa, 8-17 September 2003.
(Sacramento : InterEnvironment, California Institute
of Public Affairs, 2005).

Vermeij, G. (1987). ‘Evolution and Escalation: An
Ecological History of Life’. Princeton: Princeton
University Press.

Whitehouse, J. F. (1990a). Conserving What? The Basis of
Nature Conservation Reserves inNew South Wales
1 967- 1 989. Australian Zoologist 26 , 11-21.

Whitehouse, J. F. (1990b). The Future of Nature

Conservation Reserve Establishment Programmes.
Australian Zoologist 26 , 96-100.

Proc. Linn. Soc. N.S.W., 136, 2014

191

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

APPENDIX 1

Gallery: Linnean Society Conference (September - October 2011). All photos by Dan Lunney.

Emma Gorrod loading Paul Adam’s Royal National
Park presentation in the conference room at Kamay
Botany Bay National Park (29 September 2011).

Val Attenbrow on the Forest Path explaining the
Aboriginal use of the Park (1 October 2011).

David Keith on the Forest Path, Royal National Park
(1 October 2011).

David Keith on the heath in Royal National Park,
showing where some of his long-term plots are
located (1 October 2011).

192

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

John Pickett describing the geological attributes of a site (1 Octo-
ber 2011). What is most prominent in the photo is the clearing, and
it shows what the park would look like if it were to be cleared, or
had been cleared in the previous two centuries.

A sure sign of the continuing presence of the cryptic
Javan rusa deer ( Cervus timorensis ) in Royal Nation-
aLPark(l October 2011).

John Pickett
explaining the
geological basis
of Royal Na-
tional Park (1
October 2011).

Proc. Linn. Soc. N.S.W., 136, 2014

193

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

A felled tree on the Forest Path, part of which still remains, as does the stump. A
education notice nearby states: “Logging was permitted on at least two occa-
sions in the first quarter of this [20th] century”.

194

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Appendix 2. Parks and Reserves (pre-1967).

Park/Reserve

Modifications

Financial Year

Area (Ha)

Royal NP

1879

7284.34

2 (+)

1881

14164.00

3 (+)

1883

14171.69

4 (+)

1887

14698.18

5(+)

1900-1964

15383.72

6 (+)

1964

15769.39

7(-)

1966

14384.55

8 (-)

1967

14250.20

9 (+)

1967

14851.96

Blue Mountains NP

1960

62726.27

2 (+)

1963

68739.9

3 (+)

1966

95028.28

4 (+)

1967

98772.03

5(-)

1967

98367.34

Brisbane Water NP

1960

6070.28

2 (+)

1960

6147.98

3 (+)

1965

6181.98

4 (+)

1966

6610.94

5(+)

1967

6692.69

6 (-)

1967

6691.48

Dharug NP

1967

11748.83

Gibraltar Range NP

1963

13961.65

2 (+)

1966

15378.05

Kosciusko NP (now Kosciuszko)

1944

527019.68

2 (+)

1945

527270.59

3 (+)

1950

531438.85

4 (+)

1950

531756.93

Ku-ring-gai Chase NP

1962

14244.93

2 (-)

1962

14238.46

3 (-)

1967

14187.47

4 (+)

1967

14285.4

Morton NP

1939

18210.85

2 (+)

1963

18213.69

3 (+)

1965

18214.9

4 (-)

1965

18214.5

5(+)

1967

18240.8

Mount Kaputar NP

1960

4168.26

2 (+)

1967

14244.93

New England NP

1935

16855.16

2 (+)

1940

22520.76

3 (+)

1942

22723.1

4 (+)

1959

22724.72

5(-)

1959

22723.1

6 (-)

1967

22237.48

7 (+)

1967

22844.5

Proc. Linn. Soc. N.S.W., 136, 2014

195

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Warrumbungle NP

1962

3237.49

2 (+)

1967

6239.04

Barangary State Park

1887

849.84

2 (-)

1967

797.23

Bouddi State Park

1959

473.48

2 (+)

1959

518

3 (+)

1967

530.14

Bundanoon State Park

1961

1347.6

Dorrigo State Park

1928

1416.4

2 (-)

1930

1415.59

3 (+)

1936

1566.13

Gloucester Tops State Park

1960

1550.76

Heathcote State Park

1963

1578.27

Mount Warning State Park

1966

2116.51

Muogamarra State Park

1955

829.61

2 (+)

1962

1120.98

3 (-)

1967

1112.89

Bare Island Historic Site

1965

1.21

Captain Cook’s Landing Place

1900

95.51

2 (+)

1965

105.22

3 (+)

1967

283.28

Hill End Historic Site

1967

27.52

La Perouse Monuments Historic Site

1956

7.28

2 (+)

1967

7.69

Mootwingee Historic Site

1967

485.62

Vaucluse House Historic Site

1967

7.69

Barren Grounds Nature Reserve

1956

1489.241

2 (+)

1960

1776.5674

Bell Bird Creek Nature Reserve

1965

53.4184

Bermaguee Nature Reserve

1967

607.0275

Bird Island Nature Reserve

1960

7.2843

Black Ash Nature Reserve

1965

89.0307

Boondelbah Nature Reserve

1960

9.3078

Boorganna Nature Reserve

1955

267.0921

2 (+)

1958

308.2688

3 (+)

1962

382.7308

Bowraville Nature Reserve

1963

54.6325

2 (+)

1964

58.477

Brush Island Nature Reserve

1964

46.5388

Buddigower Nature Reserve

1964

137.5929

Cocopara Nature Reserve

1964

1308.347

2 (+)

1965

4646.998

Cook Island Nature Reserve

1960

4.6539

Coolbaggie Nature Reserve

1963

381.2133

Cudmirrah Nature Reserve

1959

125.4522

Curumbenya Nature Reserve

1965

2832.795

2 (+)

1967

8599.556

196

Proc. Linn. Soc. N.S.W., 136, 2014

D. LUNNEY

Devils Glen Nature Reserve

1965

40.4685

Five Islands Nature Reserve

1960

26.7092

Georges Creek Nature Reserve

1968

1189.774

Goonawarra Nature Reserve

1967

437.0598

Goura Nature Reserve

1967

390.521

Gurumbi Nature Reserve

1956

151.757

Illawong Nature Reserve

1964

50.5856

John Gould Nature Reserve

1955

26.3045

Julian Rocks Nature Reserve

1961

0.4047

Limpinwood Nature Reserve

1963

2321.273

2 (+)

1967

2442.6785

Lion Island Nature Reserve

1956

8.0937

Little Broughton Island Nature
Reserve

1961

36.4217

Macquarie Nature Reserve

1966

2.4477

Manobalai Nature Reserve

1968

2913.732

Moon Island Nature Reserve

1960

1.0117

Mount Seaview Nature Reserve

1965

194.2488

Munghorn Gap Nature Reserve

1961

2853.029

2 (+)

1968

2994.6688

Muogamarra Nature Reserve

1960

303.5147

2 (+)

1965

801.2788

3 (+)

1967

803.9093

Nadgee Nature Reserve

1958

11331.18

2 (+)

1961

11655.9397

3 (+)

1966

11836.0245

Narrandera Nature Reserve

1966

72.8433

North Rock Nature Reserve

1959

4.0469

Pulletop Nature Reserve

1963

145.0796

Quanda Nature Reserve

1963

429.3708

2 (+)

1967

853.8854

Round Hill Nature Reserve

1960

5179.968

2 (+)

1964

5252.8113

3 (+)

1967

5637.2621

Rowleys Creek Gulf Nature Reserve

1962

1659

Sherwood Nature Reserve

1967

1359.742

South West Solitary Island Nature
Reserve

1961

3.2375

Split Solitary Island Nature Reserve

1961

3.6422

Tabletop Nature Reserve

1966

103.5184

The Basin Nature Reserve

1964

2272.711

The Charcoal Tank Nature Reserve

1966

86.4002

The Hole Gulf Nature Reserve

1965

737

The Rock Nature Reserve

1963

271.139

Tollgate Islands Nature Reserve

1959

12.1406

Proc. Linn. Soc. N.S.W., 136, 2014

197

ROYAL NATIONAL PARK IN HISTORICAL PERSPECTIVE

Tucki Tucki Nature Reserve

1963

1.4948

2 (+)

1964

3.2375

3 (+)

1967

4.0026

Winburndale Nature Reserve

1968

3642.165

Wongarbon Nature Reserve

1966

99.1478

Proc. Linn. Soc. N.S.W., 136, 2014

198

D. LUNNEY

Appendix 3. National Parks Estate (1968-2013) at 30 June of the financial year.

Financial Year

Area (ha)

1968

894,872

1969

960,901

1970

1,096,776

1971

1,201,814

1972

1,379,278

1973

1,626,702

1974

1,638,563

1975

1,714,789

1976

1,852,407

1977

1,917,887

1978

2,076,950

1979

2,291,591

1980

2,884,692

1981

2,975,628

1982

3,039,640

1983

3,236,999

1984

3,346,667

1985

3,368,447

1986

3,415,196

1987

3,485,124

1988

3,697,308

1989

3,811,073

1990

3,853,541

1991

3,859,959

1992

3,945,810

1993

3,951,314

1994

3,955,318

1995

4,030,559

1996

4,273,545

1997

4,536,513

1998

4,553,084

1999

5,032,553

2000

5,099,674

2001

5,387,102

2002

5,419,344

2003

5,899,882

2004

5,948,814

2005

6,092,447

2006

6,487,055

2007

6,641,256

2008

6,682,405

2009

6,725,069

2010

6,763,629

2011

7,077,769

2012

7,079,707

2013

7,083,343

Proc. Linn. Soc. N.S.W., 136, 2014

199

200

A Comparative Study of the Australian Fossil Shark Egg-Case
Palaeoxyris duni, with Comments on Affinities and Structure

Graham McLean

21 Loxton Place, Forestville, NSW 2087, Australia, (mcleangg@bigpond.com)

Published on 4 July 2014 at http://escholarship.library.usyd.edu.au/joumals/index.php/LIN

McLean, G. (2014). A comparative study of the Australian fossil shark egg-case Palaeoxyris duni, with
comments on affinities and structure. Proceedings of the Linnean Society of New South Wales 136 , 20 1 -
218.

The enigmatic fossil noted by Dun in 1913 as Spirangium and named by Crookall in 1930 as
Palaeoxyris duni is described in detail in the light of the discovery of other Palaeoxyris specimens, which
are now accepted by most workers to be shark egg-cases. Palaeoxyris duni is the only Australian shark
egg-case yet described and is one of the largest Palaeoxyris species so far discovered. Comparison of the
macro morphology of P. duni with other described Palaeoxyris specimens confirms that it is a separate
form species. The palaeoenvironment in which P. duni was deposited was a Triassic low lying fluvial and
lacustrine coastal floodplain. One of the Triassic species of hybodontid sharks was the possible egg producer
as these fishes have been shown to penetrate freshwater environments. The process of egg production in the
nidamental gland of modem sharks is applied to conjecture about the egg-case stmcture of ancient sharks.
The egg-cases of Heterodontus have a helical structure broadly similar to that of Palaeoxyris except that
Palaeoxyris have four or six bands in their constmction compared to two for the modern Heterodontus.
Evidence of shark nurseries, clustered egg-cases and tendril attachment of Palaeoxyris egg-cases indicates
ancient shark breeding behaviour was similar to that of modern oviparous sharks.

Manuscript received 1 March 2013, accepted for publication 23 July 2014.

Keywords: Beacon Hill, Brookvale, Hybodontoidea, Heterodontus , nidamental gland, Palaeoxyris , shark
egg-case, Sydney Basin, Triassic.

INTRODUCTION

Palaeoxyris was an enigmatic fossil when first
described by Brongniart ( 1 828) as a rhombic patterned
capsule with tapered ends. Plant and animal affinities
were subsequently suggested for these fossils, but
mounting evidence of their shark egg-case origin has
finally been accepted by most workers (Fischer and
Kogan 2008).

Dun (1913) briefly described four imperfect
specimens found at the Beacon Hill Quarry in
Brookvale, NSW, Australia, classifying them as
Spirangium and alluding to the possibility that they
could be either fructifications of plants or the egg-
cases of primitive selachians. Since then, work has
been carried out by Crookall (1928, 1930, 1932),
Brown (1950), Zidek(1976), Bottcher (2010), Fischer
et al. (2010, 2011, 2013) and others on specimens
found in Britain, Europe, Kyrgizstan and North
America and a considerable amount of taxonomic
data has been assembled for specimens found in

the northern hemisphere. However, apart from the
Brookvale specimens mentioned by Dun (1913)
there have been no further specimens described in
the southern hemisphere and no detailed comparative
taxonomic study has been carried out on the Brookvale
specimens.

This paper provides a detailed description of the
Brookvale specimens and compares them to other
specimens described from the northern hemisphere,
as well as discussing the palaeoenvironmental
implications and the links to extant shark behaviour
and egg-case structure.

BRIEF REVIEW OF HISTORICAL RESEARCH

A thorough historical literature review on
Palaeoxyris has previously been presented by Fischer
and Kogan (2008), but a brief summary of that paper
and other references will help put this study into
context.

FOSSIL SHARK EGG-CASE

In 1828 Brongniart was the first to describe a
rhombically patterned enigmatic fossil which he
named Palaeoxyris regularis, considering it a plant
inflorescence. During the 19 th century further similar
specimens were discovered. Three form genera were
erected {Palaeoxyris, Vetacapsula and Fayolia ) and
a number of species named. Workers continued to
allocate a plant origin to them, until Beer (1856)
compared them to a specimen tentatively identified
as an egg-case. Schenk (1867) noted their external
similarity to egg-cases of extant sharks. The rhombic
pattern on specimens was recognised as a taphonomic
effect of compressing a body with spirally wound ribs
(Quenstedt 1867; Schenk 1867). However, by the end
of the 19 th century many workers still considered the
fossils to be of plant origin.

Moysey (1910) advanced the argument for a
shark origin with a detailed morphological description
of pedicle, body and beak and the concept that
ancient sharks could enter river estuaries to breed. It
was at this time that the only Australian specimens
of Palaeoxyris were found at the Beacon Hill Quarry
in Brookvale, a northern suburb of Sydney. These
specimens were referred to the genus Spirangium in
a brief note by Dun (1913). Crookall (1928, 1930,
1932) presented a series of detailed studies of the
morphology and affinities of the three form genera,
drawing on specimens from the Carboniferous Coal
Measures of Britain and Europe, and named many
new species, including the Australian specimens
which he named Palaeoxyris duni. Crookall
(1932) rejected a plant origin for these genera and
advocated the elasmobranch egg-case hypothesis.
After CrookalTs thorough analyses, discussion
turned to the most likely producer of the eggs. Both
xenacanthid and hybodontid sharks were suggested,
and palaeoecological studies were carried out to link
shark remains with the egg-case sites (Zidek 1976).
Additional specimens were described from North
America (Brown 1950; Zidek 1976).

By the beginning of the 2 1 st century new evidence
pointed to hybodontid sharks being the producers
of Palaeoxyris and xenacanthids being producers
of Fayolia (Fischer and Kogan 2008), whereas the
producer of Vetacapsula has been attributed to the
holocephalans (Fischer et al. 2013). Elasmobranch
egg-cases were found in Kyrgyzstan (Fischer et al.
2007), Triassic Palaeoxyris have been found in North
America (Fischer et al. 2010) and Germany (Bottcher
2010), and Triassic juvenile shark teeth microfossils
have been discovered in association with Palaeoxyris
in Kyrgyzstan (Fischer et al. 2011). Fischer et al.
(2013) carried out a phylogenetic analysis of the
morphology of ancient and modern chondrichthyan

egg-cases as a step towards resolving the identity of
the egg producers.

GEOLOGY AND PALAEOENVIRONMENT

Palaeoxyris duni was found within a shale
lens embedded in the Middle Triassic Hawkesbury
Sandstone of the Sydney Basin (Fig. la,b). This
sandstone was probably deposited on a vast coastal
floodplain that lay close to sea level and contained
braided rivers, scour channels, sand dunes and lakes
(Conaghan 1980). The shale lenses were formed
by deposition of finely suspended sediment in low
energy basins (Conaghan 1980), which provided
ideal anaerobic conditions in which organisms could
be preserved and fossilised.

A comparison between the flora of the Late
Carboniferous and the Middle Triassic of this
area indicated that the climate had returned to
cool temperate after the glaciation of the Pennian
(Retallack 1980), even though by the Middle Triassic
the Sydney Basin was within the Antarctic Circle (the
poles were ice free during this period) (Fig. lc).

The shale lens quarried on Beacon Hill was
deposited during the Anisian Stage of the Middle
Triassic and was composed of fine grey to black
laminated mudstone about eight metres thick. It
preserved a wide range of Triassic fossils including
the bony fishes Ceratodus, Megapteriscus,
Agecephalichthys, Belichthys, Mesembroniscns,
Myriolepis , Brookvalia, Cleithrolepis, Macroaethes,
Leptogenichthys, Geitonichthys, Molybdichthys,
Phlyctaenichthys, Schizurichthys, Manlietta,
Procheirichthys, Saurichthys, Promecosemina (Wade
1932, 1933, 1935; Hutchinson 1973,1975), the
temnospondyl P arotosuchns brookvalensis (Watson
1 95 8 ; Welles and Cosgriff 1 965), insects Clatrotitan, Ch
oristopanorpa, Austroidelia, Mesacredites, Proha glia,
Fletcheriana, Mesonotoperla, Triassocytinopsis,
Beaconella, Triassodoecns (Tilly ard 1925;McKeown
1937; Riek 1950, 1954; Evans 1956, 1963; Bethoux
and Ross 2005), crustaceans Anaspidites, Synaustrus,
Palaeolimnadiopsis and Estheria (Chilton 1929;
Brooks 1962; Riek 1964, 1968; Webb 1978), the
xiphosurian Austrolimulus fletcheri (Riek 1955,
1968), the mollusc Protovirgus brookvalensis
(Hocknull 2000) and plants Lepidopteris, Dicroidium ,
Cladophlebis, Ginkgoites, Rissikia, Taeniopteris,
Xylopteris, Phyllotheca, Marchantites, Rienitsia,
Asterotheca, Cylostrobus (Townrow 1955; Retallack
1977, 1980, 2002; Holmes 2001). This biota points
strongly to a freshwater environment.

202

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

SYDNEY BASIN TRIASSIC STRATIGRAPHY

GROUP

FORMATION

TOPOGRAPHY

230

Bringeily

Shale

streams,
lagoons, deltas

WIANAMATTA

Minchinbury

sandstone

shoreline, alluvial

Ashfield

lagoons, back

Shale

swamps

240

HAWKESBURY

SANDSTONE

Brookvale
{ Beacon Hill)
shale lens

braided river
systems with still,
blind channels

Newport

lagoons, deltas

Bald Hill
Claystone

soils, flood plains

NARRABEEN

Bulgo

meandering

Sandstone

streams

250

claystones

flood plains, lakes
and streams.

sandstones

alluvial fans

0 °

10
20 °
30 °S

40°

50°

60°S

70°

Figure 1 - a. Sydney basin stratigraphic timeline showing the position of the Brookvale (Beacon Hill)
shale lens and the topographies during sedimentation. (Data sourced from Packham 1969; Herbert and
Helby 1980) b. Location of the Brookvale (Beacon Hill) site (modified after Damiani 1999). c. In the
Early Triassic the Sydney Basin entered the Antarctic Circle as Gondwana, containing Australia, drifted
south (modified after Hallam 1994).

INSTITUTIONAL ABBREVIATIONS

AM - Australian Museum, Sydney, New South
Wales.

BMNH - Natural History Museum, London.
MM - Geological Survey of New South Wales
(refers to Mining Museum).

SYSTEMATIC DESCRIPTION

Genus: Palaeoxyris Brongniart 1828

Type species: Palaeoxyris regular is Brongniart
1928 - Anisian, Middle Triassic. Vosges, France

Proc. Linn. Soc. N.S.W., 136, 2014

203

FOSSIL SHARK EGG-CASE

Diagnosis: (after Fischer et al. 2011:943) -
“Chondrichthyan egg capsule; three-fold division
into beak, body, and pedicle; body broadly fusiform,
gradually tapering toward each end, composed of three
or more parallel helicoidally twisted bands; anterior
end gradually tapering into shorter pointed beak;
posterior end tapering to long slender pedicle marked
by either spiral ribbing or parallel ribs; collarettes
accompanying band margins; fine longitudinal
striations on bands and collarettes; compressed
specimens with transverse rhomboidal pattern”.

Palaeoxyris duni Crookall 1930

Synonomy: Spirangium : Dun 1913, 205-206 pi. 14.
Holotype: MMF 42697a (Figs 2, 3a,b,c, 4a)

Paratype: MMF 42697b (Figs 2, 3d, 4b)

Type Horizon: Hawkesbury Sandstone Formation,
Anisian, Middle Triassic (within a shale lens).

Type Locality: Beacon Hill Quarry, Brookvale, New
South Wales, Australia.

Etymology: Named after W.S. Dun, the
palaeontologist who first presented the specimen to
the Royal Society of New South Wales on Dec 12,
1912 (published 1913).

Storage Location: The two specimens are contained
on one block which is deposited in the collection
of the Geological Survey of New South Wales at
Londonderry, New South Wales, Australia.

Description

Palaeoxyris duni is a chondrichthyan egg-case
divided into a beak, a body and a pedicle. The beak
is greater than 25 mm long and tapers to a point. The
body is fusiform and shows a spiral pattern of ribs,
and is approximately 90 mm long and 30 mm wide.
The two specimens are compressed and exhibit a
rhomboidal pattern of ribs and grooves on the body,
which is a result of the rear spiral ribs being impressed
as grooves on the front spiral pattern of ribs. The
pedicle is slightly waisted, tapers, then proceeds as a
parallel stem to its end. The pedicle is at least 90 mm
long. The body structure consists of four helical bands
with a total clockwise twist of 630 degrees from the
beak to the pedicle. The bands are an average of 7
mm wide and the twist rate forms seven segments.
The ribs formed by the longitudinal suturing of the
bands are 2 mm wide. The tapered ends of the bands
form tendrils which run parallel to each other to form
the beak and pedicle (i.e. there is no twist in the beak
or pedicle).

Remarks

Although Dun (1913) stated that he had four

Figure 2. Palaeoxyris duni holotype MMF 42697a
(left) and paratype MMF 42697b (right) on a
single slab. Scale bar 10 mm.

imperfect specimens in his possession, only the block
figured in Dun (1913) can now be located.

The single block of fine grey shale (MMF 42697)
holds two specimens, one almost complete (MMF
42967a) and one with the beak and a section of the
body missing (MMF 42697b) (Fig. 2). MMF 42967a
appears to have been abraded after discovery and has
lost some of its relief. MMF 42967b retains more
structural detail. They are compressed specimens.

MMF 42697a has an incomplete beak 25 mm
long and a body 90 mm long. At the first impression
it has a pedicle 80 mm long. However, microscopic
examination of the apparent end of the pedicle shows
that the pedicle appears to be broken at this point and
bent back at an acute angle. The broken section can
be traced back for five mm, but this still may not be
the end which could be buried in the substrate. There
is a rhomboidal pattern of ribs and grooves on the
body of the specimen. The rhomboidal pattern can be
interpreted as four bands spiralling clockwise (Fig.
4c). The bands make an angle of 40 degrees with a
latitudinal line running through the centre of the body.
Each band travels around the body for 630 degrees,

204

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

Figure 3 - a. Detail of MMF 42697a beak structure. Scale bar = 7.5 mm. b. Detail of MMF 42697a
pedicle structure. Scale bar = 5 mm. c. Detail of MMF 42697a pedicle tip. Scale bar = 5 mm. d. Detail
of MMF 4267b ribs showing striae. Scale bar = 10 mm.

Proc. Linn. Soc. N.S.W., 136, 2014

205

FOSSIL SHARK EGG-CASE

Figure 4 - a. Line tracing of MMF 42697a. b. Line tracing of MMF 42697b. c. Idealised structure ofP
duni showing the bands wrapping around the body and the band terminations in the beak and pedicle.
Scale bar = 100 mm.

then tapers, forming a tendril, and runs parallel with
the others longitudinally along the beak and pedicle.
The number of segments visible on the body formed
by the spiralling bands is seven. Including the ribs, the
body is 28 mm wide at the widest point. The ribs have
a width of 2 mm. Striae running parallel to the bands
are visible on some sections of the grooves. The band

margins are defined by ribs, but there are no obvious
flanged collarette extensions from the ribs.

MMF 42697b has similar dimensions and a
similar rhomboidal pattern of ribs and grooves to
MMF 42697a, and is also composed of four bands. Its
pedicle is at least 90 mm long. The full length of the
pedicle is uncertain as it, too, may have been broken.

206

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

However, its total length matches closely the total
observable length of the pedicle (including the broken
section) of MMF 42697a. The beak and an upper
section of the body are missing. Striae are observable
in the grooves and there is a faint indication of striae
on some sections of the ribs (Fig. 3d).

Comparison with Some Other Palaeoxyris Species

From Table 1 it can be seen that P. duni has the
longest body of any of the Mesozoic taxa and the
widest body except for P. fries si. Only one species from
the Carboniferous has a longer body (P. bohemica)
and the specimens attributed to this species display
a wide range of body sizes which may indicate that
more than one species is involved.

The basic structure of all Palaeoxyris species is
made up of a number of spirally wound bands sutured
together longitudinally. Palaeozoic specimens studied
early in the 20 th century were not analysed for the
number of bands. Mesozoic specimens studied later
(e.g. Bottcher 2010; Fischer et al. 2010, 2011) were
analysed for band number and this analysis showed
that all Mesozoic species (with the one exception
of one P. humblei specimen) were determined to be
constructed with four or six bands (Table 1). P. duni is
one of four Mesozoic taxa to have four bands, whereas
another four species have six bands. The total wrap
angles of these bands around the body vary from 1 80
to 630 degrees for those species known. P. duni has
the highest total band wrap angle of 630 degrees.
This high total band wrap angle is a product of a high
wrap angle rate and a large body size. Bottcher (2010)
observed that all Palaeoxyris species have bands that
twist in a clockwise direction. Based on the premise
that the grooves in the rhombic pattern on the body
are the impressed spiral ridges from the unexposed
side of the specimen (Bottcher 2010), the bands on
P. duni twist in a clockwise direction, confonning to
this observation.

Comparison of beak and pedicle lengths is not
a strong diagnostic tool, as they are often broken,
incomplete or missing. However, in general, beak
lengths are shorter than pedicle lengths (except for
those of P. fries si, which are virtually equal). P. duni
has a longer observable pedicle than all others except
P. friessi , but its incomplete beak does not allow
length comparison.

A number of structural features noted on other
Palaeoxyris specimens are not observable on P. duni.
These are flanged collarette extensions reported on
P. alterna (Fischer et al. 2011), P. friessi (Bottcher
2010), P. humblei (Fischer et al. 2010) and on a
Mazon creek specimen (Brown 1950), and long
tendril extensions to the beak reported on P. alterna
(Fischer et al. 2011).

There are no close matches with specimens
listed in Table 1 to the combined parameters of ‘body
length’, ‘body width’, ‘band number’ and ‘total band
wrap angle’ for P. duni.

AFFINITIES AND STRUCTURE

Over the last 190 years there has been sporadic
discussion concerning the origins of Palaeoxyris.
Initially its cone-like shape with rhombic patterning
caused Brongniart (1828) and Schenk (1864) to
allocate a plant origin to these specimens. The
realisation by Schenk (1867) and Quenstedt (1867)
that the rhombic patterning could be produced
by compression of a spirally wound object led to
the comparison by Renault and Zeiller (1888) to
shark egg-cases with spiral collarettes produced by
Heterodontus sharks. Specimens were tested for plant
cell structure but none was found (Crookall 1932).
With no evidence of plant structure, opinion swung
strongly to the specimens being of shark origin
(Moysey 1910; Crookall 1932; Zidek 1976). The
palaeoenvironment in which all Palaeoxyris species
had been found is considered to be one of either
deltaic or shallow, freshwater fluvial or lacustrine
conditions (Moysey 1910; Crookall 1928; Fischer
and Kogan 2008). Ancient sharks are known to have
inhabited these environments (Patterson 1967; Rees
and Underwood 2008). In at least two instances, in
North America and Kyrgyzstan, shark remains have
been found closely associated with Palaeoxyris
specimens (Fischer et al. 2010, 2011). Fischer et al.
(2013) carried out a cladistics analysis of ancient
and modern chondrichthyan egg-cases based on
morphological traits. Their results showed the egg-
case Vetacapsula (Fig. 5b) clustered with the egg-cases
of the Chimaeridae (ratfishes), while the egg-cases
Palaeoxyris and Fayolia (Fig. 5a) were clustered next
to all egg-cases of the neoselachans (modem sharks
and rays). Egg-cases of the Heterodontidae were
positioned as the basal egg-case type morphology of
the neoselachans.

This circumstantial evidence has led to conjecture
about the actual egg producer, its breeding behaviour
and its egg-case structure.

The Egg Producer

Sharks being cartilaginous do not leave frequent
evidence of their existence in the fossil record
- teeth, fin spines and scales are the main indicators
(Kemp 1982). However, there is enough evidence to
plot the time span of the existence of possible egg
producer families. Xenacanthids (Fig.5d) appeared
in the Carboniferous (Garvey and Turner 2006;

Proc. Linn. Soc. N.S.W., 136, 2014

207

FOSSIL SHARK EGG-CASE

!"a

to“

!■

-a.

-*4.

H— •

i-h

ccr

i-i

-a.

-4^

■ — , .

O 3

-a.

n n

p p

H— •

H-h

CD^

i-i

H— •

H-h

CD^

i-i

H— •

H-h

CD^

i-i

H— •

H-h

CD^

I-S

n n

p p

H— •

H-h

CD^

I-S

H— •

H-h

i-S

§:

O 3

H— •

H-h

ccr

|-i

§-

TO x
TO TO

TO’ <*j

*T-

O 3

-a.

O 3

!"a

5 s

^04

S3 S3

*1.

■—-4

B "o
§ I-

03 a
S5 TO
v d5 O

-a.

H— •

H-h

CD^

^ >
2 - p

— t

H- ‘ •

— t

H- ‘ •

— t

H- ‘ •

2
H- ‘ •

— t

H- ‘ •

— t

H- ‘ •

— t

H- ‘ •

p p-

H- ‘ •

Q-

•— S

m ££

C/5

C^2

C/5

H- ‘ •

C^2

H- ‘ •

C/5

C^2

S2. p-

C ^

H- ‘ •

c n

H- ‘ •

C/5

H- ‘ •

H— •

H- ‘ •

C/5

H- ‘ •

r-K

i— » •

C^2

H- ‘ •

o p

r-K

H O

l-t p

•

| 4

U) H- ‘

r-K

r— K

<— K

r"K

r — K

1— ‘

-p

U> -P OS CT* OO

-P

-p

o O

p ft

3 3

p P

§ ft

O P'
§: oo

H-k

-4

— »
H- ‘ •

— >
H- ‘ •

C/5

C^2

C/5

C^2

, — s

cro

•

"H.

H— ^

H- ‘ •

-4

H- ‘ •

C^3

IT!

C^2

' '

p’

cro

"H.

-P

O a

©\

H-4

3“

H- ‘ •

<— h

<— r

H— ^

C/5

r— h

f— K

P“

£T

— t

p’

— t

| 4

r—h

f— K

1 4

H-A

H-a

. 1

■ ‘

ft*

JP$

-a.

c§

H- k

-i^

oo o

r — K

<— K

»— K

r — K

<— K

r-K

| 4

-1^

-t-

-1^

n n

«— K

r dd
a o
g o

g n<

<— K

-p

n n

><! re
2 a

0 S

1 S
s &*

? a

*8 S'

-5

cr h
re as
re g“

3 s-

3 r

!©

S’

S«.

»5

* %

? &'

sr a
© ...

-o =

as in

^ ©

cr hj

n re in a
n g cr n

Si o »

3 B

cr©

& H

©■ a-

-5

^ 2»

!? ?

1.1
as ©
a ©

C-
m ^

S3 S3

3 3

S3 _

S* as a- _

m rP —

?d « n-

^4 3 B

a 1

3
©

^■

H ^
cr m

5c* 2

« 3
as —

1 =

||
3 ©
a a

p? ©

as 3
« re

— ©

©’

m ^

^ © O'
3 ^ re

3 cr a

&- © ©

3 n ©
© 3 2.

2 ^ § a

3 -• as

2 a O-

m 2 ft
m s cr

208

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

Figure 5 - a. Carboniferous Fayolia crenulata , BMNH V12057, part and counterpart, attributed to
xenacanthid sharks, showing diagnostic scar-lines parallel to the collarette, b. Carboniferous Vetacap-
sula cooperi , BMNH V12058, tentatively attributed to the Holocephali. (Photos of shark egg-cases by
courtesy of the Natural History Museum, London ©) c. Triassic Xenacanthus (Pleur acanthus) parvidens ,
MMF 13430, from St. Peters Brickpits, Sydney, Australia (photo courtesy of Geological Survey of New
South Wales), d. The xenacanthid shark Xenacanthus sessilis (after Schaeffer and Williams 1977:297, by
permission of the Oxford University Press).

Turner and Burrow 2011). Well preserved articulated
xenacanthid fossils have been discovered in the
Middle Triassic rocks of the Sydney Basin at St Peters
Brickpits (Woodward 1908) (Fig. 5c), at Picton (an
as yet undescribed specimen held in the Australian
Museum - AM F 137124) and at Somersby Quarry
(two as yet undescribed specimens - AM F 78948,

AM F 78958 and their counterparts) (pers. comm. S.
Turner, Queensland Museum). They died out by the
Late Triassic (Kemp 1982), whereas the hybodontids
(Fig.6e) appeared in the Carboniferous and became
extinct by the end of the Cretaceous (Springer and
Gold 1 989). By matching the span of Palaeoxyris ages
with the family life spans of sharks, some workers

Proc. Linn. Soc. N.S.W., 136, 2014

209

FOSSIL SHARK EGG-CASE

Figure 6 - a. Carboniferous Palaeoxyris carbonaria , BMNH V1173, part and counterpart, showing
rhombic impressions on body and pedicle, b. Another Palaeoxyris carbonaria, also registered as BMNH
V1173, part and counterpart, showing partly uncompressed banding on left specimen, c. Palaeoxyris
carbonaria , BMNH V12928, part and counterpart, showing uncompressed spiral banding, d. Clustered
group of Cretaceous Palaeoxyris ( Spirangium ) jugleri, BMNH 38856, with joined beaks. (Photos of shark
egg-cases by courtesy of the Natural History Museum, London ©). e. The hybodontid shark Hybodus
(after Schaeffer and Williams 1977:300, by permission of the Oxford University Press).

210

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

(Crookall 1932; Zidek 1976; Bottcher 2010; Fischer
et al. 2010, 2011, 2013) have proposed hybodontid
sharks as the producers of Palaeoxyris. The earliest
Palaeoxyris species (e.g. Figs 6a,b,c) were found in
the Carboniferous and the latest in the Cretaceous (e.g.
Fig.5d), the most recent specimens being discovered
in the Wealden Group of the Lower Cretaceous rocks
near Hastings, England.

Hybodontids grew to about two metres in length,
had an amphistylic jaw and aterminal mouth (Springer
and Gold 1989). Their two dorsal fins each contained
a spine at the leading edge, and they had heterodont
dentition (piercing and crushing) which allowed a
range of food options such as fish, crustaceans and
molluscs (Springer and Gold 1989). Claspers were
present on the male (Springer and Gold 1989) which
confirmed they practiced internal fertilisation.

Hybodontid teeth have been found in deposits
interpreted as originating in estuarine and river
palaeoenvironments (Patterson 1967; Rees and
Underwood 2006, 2008) and oxygen and strontium
isotopic analyses of juvenile teeth found in lacustrine
sediments in Kyrgyzstan have confirmed that the
young sharks had developed in fresh water (Fischer et
al. 2011). Some hybodontids therefore appear to have
inhabited brackish and freshwater environments, at
least to breed.

Modern Shark Breeding Behaviour

Many workers have noted similarities between
modern Heterodontidae egg-cases and Palaeoxyris
(Moysey 1910; Brown 1950; Zidek 1976; Bottcher
2010; Fischer et al. 2010). Heterodontidae is a family
of extant oviparous sharks that produce egg-cases
with helical ribs (in the form of collarettes). Fossil
evidence of this family has been found in the early
Miocene sediments of Victoria (Kemp 1982; Long
and Turner 1984). They have an external spine on
the leading edge of each dorsal fin and crushing
toothplates suitable for a diet of molluscs. They breed
in marine waters (O’ Gower 1995).

The egg-case of the Port Jackson Shark,
Heterodontus portusjacksoni, is constructed of two
spiral bands of collagenous material approximately
0.25 mm thick that are overlapped and sutured
longitudinally. The overlapping along the sutures
forms the collarettes (Figs 7a,b). The egg case is
cone shaped with a vent at the larger (anterior) end
which opens a few weeks after deposition allowing
the circulation of water through the egg during
incubation. Finally the young shark escapes fully
formed through this vent at the larger end, leaving a
durable, empty egg-case.

The egg-case of the Crested Horn Shark,
Heterodontus galeatus, is similar to H. portusjacksoni
(Whitley 1940), but has two long tendrils that are
extensions of the collarettes. These tendrils are used
to anchor the egg-case to algae (Fig. 7c).

Tagged Heterodontus portusjacksoni has been
tracked from Cape Naturaliste, north-east Tasmania
(latitude 41 °S) to Sydney (latitude 38°S), a distance
of 850 km, during an annual migration cycle to lay
eggs in specific sites, thus exhibiting breeding fidelity
as well as spatial memory of long migration routes
(O’ Gower 1995). Females have been observed
carrying an egg-case in their mouth and egg-cases
have been found pushed into crevices so that the
collarettes hold the egg-case firmly in place (Springer
and Gold 1989; O’Gower 1995) (Fig. 7d).

The pattern of modern shark breeding behaviour,
particularly that of the oviparous sharks such as the
Heterodontus , leads to speculation about similar
ancient shark behaviour, particularly relating to
migration, breeding fidelity and the finding of ancient
shark egg-cases in consistently similar fluvial and
lacustrine environments around the world.

Modern Shark Egg-Case Structure

Knight et al. (1996) described in detail the
macro structure, biochemistry and microstructure
of selachian egg-case formation in the nidamental
(or shell) gland of oviparous sharks. Briefly, they
explained that the nidamental gland lies in line with
the oviduct (Fig. 8a). The anterior end of the gland
faces the ostium, which is the source of the fertilised
ovum. The gland in recent species is composed of
two similar halves surrounding a lumen. Each half
works in parallel to extrude a complex collagenous
lamellated sheet along its internal surface from the
anterior zone (Fig 8b). The extruding sheets are fed
by material secreted by tubules through a row of
spinnerets, and a jelly is secreted between the sheets
to divide and “inflate” the egg-case within the lumen.
As the two parallel sheets progress down the gland
the fertilised ovum enters the anterior end of the
gland and is held between the fonning sheets. The
sheets continue forming around and past the ovum
and finally join together to provide full encapsulation.
During the extrusion of the sheets that form the two
enclosing walls of the egg-case, special rib material is
also secreted to “glue” the lateral edges of the laminar
sheets together to form lateral ribs. In egg-cases that
develop horns or tendrils (which are extensions of
the lateral ribs), the posterior horns or tendrils fonn
initially and the anterior ones fonn as the very end
of the process. The final result for almost all recent

Proc. Linn. Soc. N.S.W., 136, 2014

211

FOSSIL SHARK EGG-CASE

Figure 7 - a. Egg-case of Heterodontus portusjacksoni specimen AMS IB.673. Note the striae on the col-
larettes and bands, b. Sectioned egg-case of Heterodontus portusjacksoni specimen AMS 1.30753-002. c.
Egg-case of the Crested Horn Shark Heterodontus galeatus showing its long tendrils attached to marine
algae, d. An Heterodontus portusjacksoni carries an egg for safe placement in a crevice. (Photos a,b
courtesy of the Australian Museum, Sydney. Photo c courtesy of Mark McGrouther at the Australian
Museum, Sydney. Photo d courtesy of Jayne Jenkins).

212

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

Figure 8 - Modern sharks and egg-cases, a. Simplified diagram of a modern shark reproductive sys-
tem (modified after Springer and Gold 1989:68). b. Simplified diagram of a section through an active
nidamental gland of a modern shark - i. resting gland, ii. formation of the posterior section of the egg
case wall with fertilised ovum ready to enter the lumen, iii. the fertilised ovum enters the lumen, iv.
production of the walls of the egg case continues around and behind the ovum forming the anterior end
of the egg-case and sealing it. (modified after Knight et al. 1996:98). c. Egg-case of an Atelomycterus (a
catshark) from the China Sea. d. Egg-case of an unknown species of catshark from the Timor Sea. e.
Egg-case of Zearaja nasuta (a skate) from New Zealand. (Egg-case drawings after Whitley 1940:42,44).

egg-cases is a subrectangular structure containing
the ovum and comprising two curved sheets sealed
at the two lateral margins by ribs and sealed at the
posterior and anterior ends, with tendrils or horns

protruding from the four corners (Figs 8c,d,e). After
many months (in the case of the genus Heterodontus
between 9 to 12 months (Springer and Gold 1989))
the hatching fish finally escapes through the anterior

Proc. Linn. Soc. N.S.W., 136, 2014

213

FOSSIL SHARK EGG-CASE

end of the egg-case after the anterior seal has opened
into a slit.

However, the egg-cases of the genus H eterodontns
have a strikingly different shape - that of a helix. At
first sight it appears very different to that of all other
recent oviparous sharks. Knight et al. (1996) point
out that if the ribs are flattened and the egg-case is
twisted during formation, the above process will
produce the egg-case of the Heterodontus complete
with collarettes which are the flattened ribs. Thus a
bifurcated nidamental gland can produce a spiral egg-
case with two parallel bands. Striae are observable
on H. portnsjacksoni egg-case collarettes, possibly
due to the extrusion process during formation (Figs
7a,b). Understanding the process of formation of the
modern egg-case with two bands has implications for
the study of fossil helical egg-cases with four or six
bands.

DISCUSSION

Morphology

The macromorphology of MMF 42697a and
MMF 42697b conforms to the diagnosis of Palaeoxyris
thus confirming CrookalFs decision. As the pedicle
is not twisted this confirms previous observations by
Fischer and Kogan (2008) and Bottcher (2010) that
all Mesozoic Palaeoxyris have pedicles constructed
with tendrils laid parallel longitudinally, whereas
Carboniferous species have twisted pedicles which,
when compressed, exhibit rhomboidal patterning
(Figs 6a,b,c).

Comparison with other Palaeoxyris species
(e.g. Table 1) indicates that the body length of P.
duni is only matched or exceeded by one specimen
of P. bohemica and P. trispirilis which are both Late
Carboniferous in age. Although some extant sharks
such as the catshark Scyliorhinus canicnla produce
an intra-species range of egg-case sizes which is
determined by the size of the female and its habitat
(Springer and Gold 1989), the range is still limited
and it is therefore reasonable to conjecture that the
size of the egg-case of P. duni indicates that the egg
producer was one of the comparatively larger Triassic
hybodontids.

The combination of the P. duni macro morphology
parameters of ‘body length’, ‘body width’, ‘band
number’ and ‘total band wrap angle’ is unique and
therefore justifies its classification as a definite form
species.

A particular feature of the modern Heterodontus
egg-case is the wide flanged collarette (Figs 7a, b).
Flanged collarettes have been detected on Palaeozoic

and Mesozoic Palaeoxyris specimens (Brown 1950;
Bottcher 2010; Fischer et al. 2010, 2011), although
they are not seen on P. duni. The Carboniferous P.
helictoroides exhibits a wide/narrow pattern of
segments. The narrow segments could possibly be a
collarette impression. The Heterodontus collarette is
thin (0.25 mm) and friable when dry. It is possible that
many more ancient egg-cases might have had flanged
collarettes but that these were destroyed during the
taphonomic process.

Striae running longitudinally parallel span the
bands and collarettes of Heterodontus egg-cases
(Figs 7a, b). It is likely that these are produced by
the extrusion process within the nidamental gland
by the array of spinnerets that form the bands. Striae
are observable in many Palaeoxyris specimens
(Crookall 1932; Fischer et al. 2010) including P. duni,
particularly in the sheltered regions like the grooves.
Striae are thus strong circumstantial evidence that
Palaeoxyris had a similar egg-case formation process
to that of the modern shark genus Heterodontus.

Palaeoenvironment

The interpreted palaeoenvironment in which
P. duni was produced bears a close resemblance to
that described for many other northern hemisphere
species. The eggs were laid in a still, shallow,
freshwater lacustrine or lagoonal environment, most
likely accessible from the sea. Fossils recovered from
the fine grained shale lens in which P. duni was found
(Dun 1913) are a close match to those found with
other Triassic Palaeoxyris specimens, for example the
plants Taeniopteris, Cladophlebis and horsetails, and
invertebrates such as conchostrachans and brackish
water bivalves (Bottcher 20 1 0). A similar environment
was described for P. alterna (Fischer et al. 2011), P.
friessi (Bottcher 2010) and P. humblei (Fischer et al.
2010). Carboniferous species described by Crookall
(1928, 1930, 1932) were found in the British Coal
Measures that formed in freshwater swamps. Fischer
et al. (20 1 1 ) postulated that the producers of P. alterna
might have lived as adults in an enclosed freshwater
lake. Patterson (1967) and Rees and Underwood
(2008) conjectured that hybodonts, already capable
of travelling up rivers and lakes to breed, might have
radiated and diversified within wholly freshwater
environments under pressure of the developing marine
neoselachians in the early Jurassic and Cretaceous.

Hybodontid Sharks in Eastern Australia

There is scant evidence for the presence of
hybodontids along the coast of eastern Australia in
the Triassic and Jurassic. Woodward (1890) described
a selachian with two dorsal fins complete with spines

214

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

found in the Narrabeen Group of the freshwater
Triassic sediments of Gosford, NSW. (Unfortunately,
that specimen has not been traced (pers. comm. Susan
Turner 2012)). Another eastern Australian Triassic
hybodontid is in the process of being described (pers.
comm. Susan Turner 2012). There is also a Jurassic
specimen from the freshwater deposits of Talbragar,
NSW yet to be described (Turner et al. 2009).

Shark Behaviour

Heterodontns portnsjacksoni has been shown to
migrate long distances to feeding sites, and to return
regularly to known, shallow water marine breeding
sites (O’Gower 1995). The palaeogeographic position
of the Sydney Basin in the Triassic was within the
Antarctic Circle (Hallam 1994), meaning that during
winter there was probably a long, unbroken period of
darkness. During this darkness it was likely that food
sources for the hybodontids would either migrate
north or seasonally reduce in numbers (as do modem
krill). This would force hybodontids to migrate north
in winter. It is therefore likely that hybodontids in
these latitudes followed an annual migration pattern
of northern migration in winter, then a return to
known breeding sites in the rivers and lakes of the
Sydney Basin in summer.

Modern oviparous sharks, such as Heterodontns ,
have been observed to gather at common shark
nurseries to lay their eggs (O’ Gower 1995). It is an
advantage to a marine species that produces only a
few eggs to secure them in a safe place and protect
them from predation, random current transport
and storms. H. portnsjacksoni does this by pushing
them into rocky crevices (Springer and Gold 1989;
O’Gower 1995) (Fig.7d), H. galeatus anchors its eggs
to marine algae using long, flexible tendrils (Fig. 7c).
A flexible tendril has been discovered protruding
from beak of one Palaeoxyris alterna specimen
(Fischer et al. 2011). Fischer et al. (2010) noted
the finding of a Palaeoxyris specimen from Mazon
Creek, attached to wood fragments by beak tendrils.
Crookall ( 1 932) described and figured five P. jngleri
jointly attached by their beaks (Fig. 6d). MMF 42697
shows the beaks of the two P. dnni oriented in the
same direction, which indicates they may have been
joined or jointly anchored by their beaks. Fischer et
al. (2011) describe the finding of 31 specimens of
P. alterna (some fragmentary) in association with
juvenile shark teeth. This circumstantial evidence
links ancient shark breeding behaviour to modern
shark nursery breeding habits.

Egg-Case Formation

Fischer et al. (2013) identified nine ancient and
modern morphotypes of chondrichthyan egg-cases,

seven of which appeared in the fossil record. They
carried out a cladistics analysis of 11 taxa, based
on 15 morphological characters, which clustered
Palaeoxyris and Fayolia and grouped this cluster
next to neoselachan egg-cases.

Considering the process by which the nidamental
gland produces a shark egg-case leads to some valuable
insights into the morphology of ancient shark egg-
cases. Although the shapes of extant shark, ray and
skate egg-cases (except for the genus Heterodontns )
at first sight appear quite different from the helically
twisted Palaeoxyris (Figs 8c,d,e), they are in fact all
variations on a fundamental structure. This structure
comprises extruded posterior tendrils, enclosing
sheets (or bands), the longitudinal suturing of these
bands together, the sealing of the ends and the final
production of anterior tendrils. The egg-case of each
species varies in the size, the number or absence of
pairs of tendrils, the shape of the end seals and the
size and shape of the longitudinal sutures. In the case
of the genus Heterodontns the complete structure
is twisted into a helix. All extant sharks, rays and
skates produce egg-cases with two enclosing sheets
(or bands). The morphology of Palaeoxyris reveals
all the same elements of the fundamental structure
- sheets (or bands) longitudinally sutured forming
ribs, twisted into a helix, and a beak and pedicle
formed by the joining of tendrils which each originate
at the end of a rib. Palaeoxyris species, however,
have four or six bands. This leads to the conjecture
that ancient nidamental glands were divided into
four or six parallel sections, which each extruded a
separate band. This in turn leads to the conclusion
that the combination of egg-case body size (within a
tolerance), the number of bands and the helical twist
rate would identify separate egg producer species,
as each egg producer species would have a common
nidamental gland structure.

Diagnostic Parameters

Based on the premise that each egg producing
species would have a common nidamental gland
structure, for the reasons set out above, the number
of bands and the helical twist rate for each egg-case
form species would be diagnostic, coupled with body
size. Bottcher (2010) commented that all Mesozoic
Palaeoxyris specimens so far described had even
numbers of bands (either four or six or even greater).
Recent egg laying sharks and rays all have two bands,
thus supporting the concept that the fossil egg-cases
were produced by a different clade, such as the
hybodonts.

However, there is one recent paper that tests this
concept. Fischer et al. (2010) reported the finding
of three specimens of P. hnmblei , two with four

Proc. Linn. Soc. N.S.W., 136, 2014

215

FOSSIL SHARK EGG-CASE

bands and one with three bands. Due to taphonomic
distortion it is often difficult to determine the exact
number of bands (Bottcher 2010), particularly if the
specimens are compressed and the edges of the body
are not well defined or are buried in the substrate. If
further evidence of variation of band number within
form species is found, doubt may be thrown on band
number as a significant diagnostic parameter.

CONCLUSIONS

The basic morphology of Palaeoxyris duni
shows a strong relationship to northern hemisphere
Mesozoic Palaeoxyris species, but the essential
diagnostic indicators of ‘body size’, ‘band number’
and ‘total wrap angle’ in combination do not match
other specimens, confirming that P. duni is a separate
form species.

The existence of common structures, such
as helical bands, collarettes, tendrils and striae in
Palaeoxyris form species and modern oviparous
shark egg-cases is convincing evidence that ancient
sharks produced Palaeoxyris. Geographic, temporal
and environmental constraints lead to the initial
consideration that either xenacanthid or hybodontid
sharks produced Palaeoxyris species. Currently the
weight of circumstantial evidence favours the family
Hybodontidae. If hybodontid species were the egg-
case producers, specimen MMF 42697 is further
evidence of their presence on the eastern coast of
Australia during the mid-Triassic along with the
specimens known from the Sydney Basin.

It is probable that the nidamental glands of
hybodont shark species produced egg cases in a
similar manner to modern oviparous sharks, except
that the glands were divided into more than two
parallel sections. Thus the diagnostic features that
define a Palaeoxyris form species by association
define a shark species.

The palaeoenvironment in which P. duni was
deposited matches that described for most, if not
all, other Palaeoxyris species. This is essentially a
deltaic or shallow, freshwater, fluvial, lacustrine or
lagoonal coastal environment accessible from the
sea. A significant body of evidence for hybodontid
movement into freshwater systems, particularly
the finding of fossil teeth and fin spines, shows that
these sharks were capable of making the transition
from marine to freshwater, if only to breed. There are
indications that breeding habits such as the formation
of shark nurseries and egg-case attachment were
practiced by ancient sharks in a similar manner to
those of modern sharks.

As the Sydney Basin where the P. duni specimens
were found was within the Antarctic Circle during the
Middle Triassic Period, it is likely that hybodontids
followed an annual migration pattern of northward
winter movement and a return to regular favoured
breeding areas to the south during summer for
breeding purposes.

Most workers accept the hypothesis that
Palaeoxyris are shark egg-cases. There are still some
questions to be answered. The search is on for any
fossil egg-cases, including Palaeoxyris, containing
embryonic shark remains.

ACKNOWLEDGEMENTS

My thanks go to Yong Yi Zhen (Geological Survey
of New South Wales) who has been a friend and mentor
over the last eight years and encouraged me to write this
paper. I also thank Robert Jones (Australian Museum) for
his constructive criticism that guided the honing of my
arguments and for his help in preparations of specimens
for photography. Ian Percival (Geological Survey of New
South Wales) furnished me with the key specimen and was
just as excited as I was when we tracked down W.S. Dun’s
enigmatic Palaeoxyris fossil in the Geological Survey of
New South Wales collection. Ian also generously gave his
time to suggest scientific and grammatical improvements to
the first draft of this paper. Mark McGrouther (Australian
Museum) contributed valuable information on modern
sharks as well as access to the Australian Museum’s
collection of Heterodontus shark egg-cases. Martha
Richter, Peta Hayes and Martin Munt (Natural History
Museum, London) made me welcome and arranged for
my examination of their Palaeoxyris specimens and
the subsequent production of photographs. Sue Turner
(Queensland Museum) willingly provided her expert
knowledge on the evidence for existence of hybodontid
sharks along the ancient eastern coast of Australia. My
thanks go to Glenn Brock (Macquarie University) whose
enthusiasm for the teaching of palaeontology spurred and
consolidated my own desire to make a contribution to this
field of earth sciences. Finally, I would like to thank the
referees and the editor who spent considerable time and
effort reviewing and improving the manuscript.

REFERENCES

Beer, J.G. (1856). Einiges uber Bromeliaceen.

Bonplandia-Zeitschrift fur die gesammte Botanik IV
24 , 382-384.

Bethoux, O. and Ross, A.J. (2005). Mesacridites Riek,

1954 (Middle Triassic; Australia) transferred from
Protorthoptera to Orthoptera: Locustavidae. Journal
of Paleontology 79 , 607-610.

Bottcher, R. (2010). Description of the shark egg

capsule Palaeoxyris friessi n. sp. from the Ladinian

216

Proc. Linn. Soc. N.S.W., 136, 2014

G. McLEAN

(Middle Triassic) of SW Germany and discussion
of all known egg capsules from the Triassic of the
Germanic Basin. Palaeodiversity 3, 123-139.

Brongniart, A. (1828). Essai d’une Flore du gres bigarre.
Annales des science naturelles 15 , 435-460.

Brooks, ELK. (1962). On the fossil Anaspidacea, with
a revision of the classification of the Syncarida.
Crustaceana 4 , 229-242.

Brown, R.W. (1950). Cretaceous fish egg capsule from
Kansas. Journal of Paleontology 24 , 594-600.

Chilton, C. (1929). Note on a fossil shrimp from the

Hawkesbury Sandstones. Journal and Proceedings of
the Royal Society of New South Wales 62 , 366-368.

Conaghan, P.J. (1980). The Hawkesbury Sandstone: gross
characteristics and depositional environment. In
‘Bulletin No. 26 - A Guide to Sydney Basin’ (Eds.

C. Herbert and R. Helby), 188-253. (Department
of Mineral Resources - Geological Survey of New
South Wales: Sydney, Australia).

Crookall, R. (1928). XIII. -Palaeozoic species of

Vetacapsula and Palaeoxyris. Summary of Progress
of the Geological Survey of Great Britain and the
Museum of Practical Geology for the Year 1927 Part
II, 87-107.

Crookall, R. (1930). II.-Further morphological studies
in Palaeoxyris , etc. Summary of Progress of the
Geological Survey of Great Britain and the Museum
of Practical Geology for the Year 1929 Part 111, 8-36.

Crookall, R. (1932). IX.-The nature and affinities of
Palaeoxyris , etc. Summary of Progress of the
Geological Survey of Great Britain and the Museum
of Practical Geology for the Year 1931 Part II, 122-
140.

Damiani, R.J. (1999). Giant temnospondyl amphibians
from the Early to Middle Triassic Narrabeen Group
of the Sydney Basin, New South Wales, Australia.
Alcheringa 23, 87-109.

Dun, W.S. (1913). Note on the occurrence of the genus
Spirangium in the Hawkesbury Series of New South
Wales. Journal and Proceedings of the Royal Society
of New South Wales 45, 205-206.

Evans, J.W. (1956). Palaeozoic and Mesozoic Hemiptera
(Insecta). Australian Journal of Zoology 4 , 165-258.

Evans, J.W. (1963). The systematic position of the
Ipsviciidae (Upper Triassic Hemiptera) and some
new Upper Permian and Middle Triassic Hemiptera
from Australia (Insecta). Journal of the Entomology
Society of Queensland 2 , 17-23.

Fischer, J., Axsmith, B.J. and Ash, S.R. (2010). First
unequivocal record of the hybodont shark egg
capsule Palaeoxyris in the Mesozoic of North
America. Neues Jahrbuch fur Geologie und
Pdlaontologie Abhandlungen 255 , 327-344.

Fischer, J. and Kogan, I. (2008). Elasmobranch egg
capsules Palaeoxyris, Fayolia and Vetacapsula as
subject of palaeontological research - an annotated
bibliography. Freiberger For schungshefte 528 , 75-
91.

Fischer, J., Licht, M., Kriwet, J., Schneider, J.W.,

Buchwitz, M. and Bartsch, P. (2013). Egg capsule

morphology provides new information about the
interrelationships of chondrichthyan fishes. Journal
of Systematic Palaeontology, DOE 10.1080/1477201
9.2012.762061, 1-11.

Fischer, J., Voigt, S. and Buchwitz, M. (2007). First
elasmobranch egg capsules from freshwater lake
deposits of the Madygen Formation (Middle to Late
Triassic, Kyrgyzstan, Central Asia). Freiberger
For schungshefte 524 , 41-46.

Fischer, J., Voigt, S., Schneider, J.W., Buchwitz, M. and
Voigt, S. (2011). A selachian freshwater fauna from
the Triassic of Kyrgyzstan and its implication for
Mesozoic shark nurseries. Journal of Vertebrate
Paleontology 31 , 937-953.

Garvey, J.M. and Turner, S. (2006). Vertebrate
microremains from the presumed earliest
Carboniferous of the Mansfield Basin, Victoria.
Alcheringa 30 , 43-62.

Hallam, A. (1994). An outline of Phanerozoic

biogeography. (Oxford University Press: United
Kingdom).

Herbert, C. and Helby, R. (1980) (eds). A Guide to the
Sydney Basin. Geological Survey of New South
Wales Bulletin 26 , 1-603.

Hocknull, S.A. (2000). Mesozoic freshwater and estuarine
bivalves from Australia. Memoirs of the Queensland
Museum 45 , 405-426.

Holmes, W.B. (2001). Equisetalean plant remains from the
Early to Middle Triassic of NSW, Australia. Records
of the Australian Museum 53 , 9-20.

Hutchinson, P. (1973). A revision of the perleidiform
fishes from the Triassic of Bekker’s Kraal (South
Africa) and Brookvale (New South Wales). Bulletin
of the British Museum (Natural History) Geology 22 ,
235-354.

Hutchinson, P. (1975). Two Triassic fish from South Africa
and Australia, with comments on the evolution of
Chondrostei. Palaeontology 18 , 613-629.

Kemp, N.R. (1982). Chondrichthyans in the Tertiary
of Australia. In ‘The Fossil Vertebrate Record of
Australia’ (Eds. P.V. Rich and E.M. Thompson),
88-117. (Monash University Offset Printing Unit:
Clayton, Victoria, Australia).

Knight, D.P, Feng, D. and Stewart, M. (1996). Structure
and function of the selachian egg case. Biological
Reviews of the Cambridge Philosophical Society 71 ,
81-111.

Long, J. and Turner, S. (1984). A checklist and

bibliography of Australian fossil fish. In ‘Vertebrate
Zoogeography and Evolution in Australia. Animals
in Space and Time’ (Eds. M. Archer and G. Clayton),
235-254. (Hesperian Press: Carlisle, Western
Australia).

McKeown, K.C. (1937). New fossil insect wings

(Protohemiptera, family Mesotitanidae). Records of
the Australian Museum 20 , 31-37.

Moysey, L. (1910). On Palaeoxyris and other fossils
from the Derbyshire and Nottinghamshire coalfield.
The Quarterly Journal of the Geological Society of
London 66, 329-345.

Proc. Linn. Soc. N.S.W., 136, 2014

217

FOSSIL SHARK EGG-CASE

O’Gower, A.K. (1995). Speculations on a spatial

memory for the Port Jackson Shark ( Heterodontus
portjacksoni) (Meyer) (Heterodontidae). Marine
Freshwater Research 46 , 861-871.

Packham, G.H. (1969) (ed.). The Geology of New South
Wales. Journal of the Geological Society of Australia

16 , 1-653.

Patterson, C. (1967). British Wealden sharks. Bulletin of
the British Museum (Natural History): Geology 11 ,
283-350.

Quenstedt, A. (1867). ‘Handbuch der Petrefaktenkunde’
Zweite umgearbeitete und vermehrte Auflage p982.
(H. Laupp’sche Buchhandlung: Tubingen).

Rees, J. and Underwood, C.J. (2006). Hybodont sharks
from the Middle Jurassic of the Inner Hebrides,
Scotland. Transactions of the Royal Society of
Edinburgh: Earth Sciences 96 , 351-363.

Rees, J. and Underwood, C.J. (2008). Hybodont sharks
of the English Bathonian and Callovian (Middle
Jurassic). Palaeontology 51 , 117-147.

Renault, B. and Zeiller, R. (1888). Sur Tattribution de
genres Fayolia et Palaeoxyris. Comptes rendus
hebdomadaires des seances de l Academie des
sciences 107 , 1022-1025.

Retallack, G. (1977). Reconstructing Triassic vegetation
of eastern Australasia: a new approach for the
biostratigraphy of Gondwanaland. Alcheringa 1 , 247-
277.

Retallack, G. (1980). Late Carboniferous to Middle

Triassic megafossil floras from the Sydney Basin. In
‘Bulletin No. 26 - A Guide to Sydney Basin’ (eds.

C. Herbert and R. Helby), 385-430. (Department
of Mineral Resources - Geological Survey of New
South Wales: Sydney, Australia).

Retallack, G. (2002). Lepidopteris callipteroides, an
earliest Triassic seed fem of the Sydney Basin,
southeastern Australia. Alcheringa 26 , 475-500.

Riek, E.F. (1950). A fossil mecopteron from the Triassic
beds at Brookvale, NSW. Records of the Australian
Museum 22 , 254-256.

Riek, E.F. (1954). Further Triassic insects from Brookvale,
NSW (orders Orthoptera Saltatoria, Protorthoptera,
Perlaria). Records of the Australian Museum 23, 161-
168.

Riek, E.F. (1955). A new xiphosuran from the Triassic
sediments at Brookvale, New South Wales. Records
of the Australian Museum 23 , 281-282.

Riek, E.F. (1964). Merostromoidea (Arthropoda,
Trilobitomorpha) from the Australian Middle
Triassic. Records of the Australian Museum 26 , 327-
332.

Riek, E.F. (1968). Re-examination of two arthropod
species from the Triassic of Brookvale, New South
Wales. Records of the Australian Museum 27 , 313-
321.

Schaeffer, B. and Williams, M. (1977). Relationships of
fossil and living elasmobranchs. American Zoologist

17 , 293-302.

Schenk, A. (1867). ‘Die fossile Flora der Grenzschichten’
(Eds Keupers und Lias Frankens) pp 149, 195-198,
204-205. (C.W. Kreidel’s Verlag: Wiesbaden).

Springer, V.G. and Gold, J.P (1979). ‘Sharks in question:
the Smithsonian answer book’. (Smithsonian
Institution Press: Washington,D.C., USA).

Tillyard, R.J. (1925). Anew fossil insect wing from

Triassic beds near Dee Why, NSW. Proceedings of
the Linnean Society of New South Wales 50 , 374-377.

Townrow, J.A. (1955). On some species of Phyllotheca.
Journal and Proceedings of the Royal Society of New
South Wales 89 , 39-63.

Turner, S., Bean, L.B., Dettmann, M., McKellar, J.L.,
McLoughlin, S. and Thulborn, T. (2009). Australian
Jurassic sedimentary and fossil successions: current
work and future prospects for marine and non-marine
correlation. GFF 131 , 49-70.

Turner, S. and Burrow, C.J. (2011). A Lower

Carboniferous xenacanthiform shark from Australia.
Journal of Vertebrate Paleontology 31 , 241-257.

Wade, R.T. (1932). Preliminary note on Macroethes
brooMalei, representing a new family of
chondrostean fishes, the Pholidopleuridae. Annals
and Magazine of Natural History (Series 10) 9. 473-
475.

Wade, R.T. (1933). On a new Triassic catopterid fish from
New South Wales. Annals and Magazine of Natural
History (Series 10) 12 , 121-125.

Wade, R.T. (1935). ‘The Triassic fishes of Brookvale, New
South Wales’. (British Museum (Natural History) and
The Oxford University Press: United Kingdom).

Watson, D.M.S. (1958). Anew labyrinthodont

( Paracyclotosaurus ) from the Upper Trias of New
South Wales. Bulletin of the British Museum (Natural
History) Geology 3 , 233-263.

Webb, J.A. (1978). A new Triassic Palaeolimnadiopsis
(Crustacea: Conchostraca) from the Sydney Basin,
New South Wales. Alcheringa 2 , 261-267.

Welles, S.P and Cosgriff, J.W. (1965). A revision of
the labyrinthodont family Capitosauridae and a
description of Parotosaurus peabodyi n. sp. from
the Wupatki Member of the Moenkopi Formation
of northern Arizona. University of California
Publications in Geological Science 54 , 1-61.

Whitley, G.P (1940). ‘The fishes of Australia Part 1: The
sharks, rays, devil-fish, and other primitive fishes
of Australia and New Zealand’. (Royal Zoological
Society of New South Wales: Sydney, Australia).

Woodward, A.S. (1890). The fossil fishes of the
Hawkesbury Series at Gosford. Memoirs of
the Geological Survey of New South Wales,
Palaeontology 4, 1-55.

Woodward, A.S. (1908). The fossil fishes of the
Hawkesbury Series at St. Peter’s. Memoirs of
the Geological Survey of New South Wales,
Palaeontology 10 , 1-29.

Zidek, J. (1976). A new egg capsule from the

Pennsylvanian of Oklahoma, and remarks on the
chondrichthyan egg capsules in general. Journal of
Paleontology 50 , 907-915.

218

Proc. Linn. Soc. N.S.W., 136, 2014

Reproductive Biology of Estuarine Pufferfish, Marilyna
pleurosticta and Tetractenos hamiltoni (Teleostei:
Tetraodontidae) in Northern New South Wales: Implications for

Biomonitoring

Rumeaida Mat Piah 1 ’ 2 and Daniel J. Bucher 1 *

'Marine Ecology Research Center, Southern Cross University, P.O. Box 157, Lismore NSW, 2480, Australia 5
2 School of Fisheries and Aquaculture Sciences, University Malaysia Terengganu, 21030 Kuala Terengganu,

Malaysia. Corresponding author: daniel.bucher@scu.edu.au

Published on 22 September 2014 at http://escholarship.library.usyd.edu.au/journals/index.php/LIN

Mat Piah, R. and Bucher, D.J. (2014). Reproductive biology of estuarine pufferfish, Marilyna pleurosticta
and Tetractenos hamiltoni (Teleostei: Tetraodontidae) in northern New South Wales: implications for
biomonitoring. Proceedings of the Linnean Society of New South Wales 136 , 219-229.

Because of their broad distribution, site fidelity and long lifespan pufferfish (family Tetraodontidae)
show potential as biomonitors of estuarine water quality, and as understanding the reproductive cycle is
crucial to interpreting variations in contaminant loads in the tissues of biomonitors, we investigated the
spawning season, length at maturity and body condition of two small sympatric pufferfish species ( Marilyna
pleurosticta and Tetractenos hamiltoni), in the Richmond Estuary, NSW. M. pleurosticta spawned in spring
while T. hamiltoni spawned in winter. Female and male M. pleurosticta matured at a similar size (50%
mature at 90 mm total length). In T. hamiltoni there was a more marked difference in size at 50% maturity,
with males maturing at 80 mm and females at 110 mm TL. From the high values for hepatosomatic index
(HSI) and its strong inverse relationship with gonadosomatic index (GSI) we inferred that lipid reserves
in the liver play an important role in gonad maturation and spawning. Somatic condition factor (K ) also
varied, albeit less so, throughout the year, suggesting that body fat and muscle play lesser roles in providing
energy for reproduction. Seasonality of liver lipid content and different spawning seasons have important
implications for designing sampling strategies using these fish, especially when monitoring lipophilic
contaminants.

Manuscript received 24 May 2014, accepted for publication 17 September 2014.

Key words: banded pufferfish, biomonitoring, common pufferfish, condition, gonadosomatic index,
hepatosomatic index, Marilyna pleurosticta, reproduction, spawning, Tetractenos hamiltoni.

INTRODUCTION (Booth and Schultz 1999; Alquezar et al. 2006; Mat

Piah 2011). Despite this, there are very few studies of
Pufferfish (Family Tetraodontidae) are common the biology of estuarine pufferfish (NSW Department
in tropical, subtropical and temperate estuaries (Bell of Primary Industries 2006).

et al. 1984; Hindell and Jenkins 2004), occuring Reproductive processes may influence the

around the Australian coast and throughout the full storage, mobilisation and transfer of lipids (Merayo

range of estuarine salinities from freshwater to marine 1996; Alonso-Fernandez and Saborido-Rey 2012)

conditions (Gomon et al. 2008; Allen 2009). Whilst and hence the partitioning of lipophilic contaminants

they are potentially susceptible to being caught in among different tissues (Fletcher and King 1978),

unsustainable numbers as trawl by-catch (Stobutzki so the reproductive cycles and changes in condition

et al. 2001), few are of conservation concern in of any species to be used as a biomonitor should

Australia. Tetraodontids show potential to be used as therefore be well understood. A species on the west

biomonitors of estuarine contamination because they coast of Australia, Torquigener pleurogramma, is

are widely abundant, long-lived, appear to remain a broadcast spawner during summer and sexual

within a small home range and are carnivorous maturity is reached at two years of age (Potter et al.

REPRODUCTIVE BIOLOGY OF ESTUARINE PUFFERFISH

1988). A similar species in New Zealand Contusus
richei also matures at two years old and spawns in
summer (Habib 1979) whereas the smooth puffer
Tetractenos glaber in the Sydney region spawns in
winter (Booth and Schultz, 1999). The reproductive
biology of other tetraodontids, other than the few
large pelagic species used commercially for human
consumption (e.g. Sabrah et al. 2006) has not been
documented.

The balance between the metabolic demands for
energy and the quantity and quality of food intake
determines the amount of energy that can be stored
as lipids to fuel growth and reproduction and can
be indicated by several indices. Some commonly
used indirect measures of a fish’s energy status are
Fulton’s condition factor K, which is an index of
the body weight of a fish relative to its length and
the hepatosomatic index (HSI), which is a measure
of the relative weight of the liver, a major energy
store especially in non-fatty fish (McPherson et
al. 2011). In pufferfish, the liver is large and easily
dissected from other organs. Condition factor indices
that use gutted (somatic) weight (K) (e.g. Encina
and Granado-Lorencio 1997) can provide a better
indicator of changes in the food reserves stored in
muscle and body fat than indices using total weight,
especially in light- framed fishes such as puffers where
variations in large organs such as gonad and liver and
gut fullness can mask changes in body fat and muscle
(McPherson et al. 2011). At high K values, excess
energy results in accumulation of fat and oil reserves
and little demand on protein for energy production,
resulting in greater muscle development. At low K
values, the metabolic energy demands have depleted
lipid reserves and are supplemented by catabolism of
proteins, resulting in reduced muscle development.

In preparation for spawning HSI may decline
during gamete development as lipid reserves are
reduced (Htun-han 1978). Somatic conditionmay also
reduce in this time if somatic fat deposits or protein
are catabolised during gametogenesis or for spawning
migrations (Htun-han 1978). After spawning HSI
gradually increases as energy reserves are restored
prior to commencement of the next gametogenic cycle
(Htun-han 1978). This cycle is dependent on food
intake being sufficient to allow reserves to increase
between spawning events. If not, fecundity, egg size
or yolk content will be reduced in the subsequent
spawning (e.g. Burton 1994). The immediate post-
spawning period is also generally the time of most
rapid somatic growth (Chellapa et al. 1995).

Our study examined the reproductive cycles and
condition indices of two pufferfish species commonly
found in tropical and sub-tropical Australian estuaries,

Marilyna pleurosticta and Tetractenos hamiltoni.
These two species commonly occur from northern
New South Wales and Queensland (Grant 1987; Edgar
2000), north to Papua New Guinea (Coates 1993),
where they inhabit a broad range of habitats from
near the mouth of the estuary to its upper low-salinity
reaches. If these two widespread species were to be
used as biomonitors of estuarine water quality then
any differences between the reproductive cycles of the
species would need to be considered when choosing
the species and time of year to sample. The specific
objectives of this study were therefore to quantify the
spawning season, and length at 50% maturity of M
pleurosticta and T. hamiltoni , to determine if seasonal
variations of body condition indices occur and differ
in M .pleurosticta and T. hamiltoni and to quantify
the relationship between body condition indices and
spawning activity of the fish.

MATERIALS AND METHODS

Study area and sampling procedure

Fish were captured each month from March 2008
to March 2010 in a tributary of the Richmond River
Estuary, New South Wales, Australia (Fig. 1). Fishes
were collected as they returned from foraging over
intertidal mangrove forests by setting 12 fyke nets
with a mesh size of 12 mm, entrance radius 30 to 45
cm and wing length up to 5m in tidal channels on
the ebbing tide. Nets were set on daylight spring high
tides and retrieved as they were exposed by the falling
tide. After capture, the fish were euthanised using 1
ppt solution of Benzocaine (ethyl-p-aminobenzoate)
in water from the capture site, transported on ice
packs and frozen within 2 hours of capture.

In the laboratory, thawed fish were measured
from snout to distal edge of the caudal fin (total
length, TL to the nearest 1 mm). Total weight (TW to
the nearest O.lg) was measured using a pan balance
(Mettler Toledo PL3002). Gutted carcasses were
weighed on the same balance to provide the somatic
weight (SW). Liver and gonad tissues were removed
and weighed (LW and GW to the nearest 0.00 lg )
using an analytical balance (Mettler Toledo AL204)..

Reproductive biology analysis

Fish gender was determined using a gonad visual
census, in which testes appeared as smooth-textured
and ivory-white in colour and ovaries pink to orange
in colour with a granular texture of developing
oocytes within. Fish in which the gonads were small,
thin and transparent and unable to be confidently

220

Proc. Linn. Soc. N.S.W., 136, 2014

R.B. MAT PIAH AND D.J. BUCHER

Fig. 1. Location of study site (X) in Little Fishery Creek, a tributary of the Richmond River estuary at
Ballina, NSW (base map courtesy of D. Maher, Center for Coastal Biogeochemistry, Southern Cross
University).

Proc. Linn. Soc. N.S.W., 136, 2014

221

REPRODUCTIVE BIOLOGY OF ESTUARINE PUFFERFISH

assigned to male or female were categorised as
immature.

A total of 358 gonads from M. pleurosticta and
89 gonads from T. hamiltoni were examined. Maturity
stages were determined macroscopically and were
classified as either I, immature; II, developing; III,
spawning-capable; IV, spent/resting (= regressing/
regenerating) (modified from Brown-Peterson et
al., 2011). The spawning season was determined by
following the changes in proportions of the three non-
immature stages on a seasonal basis and by following
monthly changes in mean gonadosomatic index (GSI),
where GSI = 100 * GW/TW. Elevated GSI in mature
fish indicates that they are approaching spawning
season and a rapid reduction in GSI indicates that
spawning has recently occurred.

Length at maturity was determined as the length
at which 50% of individuals were at maturity stages
III or IV, during the spawning season. The length at
50% sexual maturity (L 50 ) was estimated by fitting
a logistic model to the combined percentage of fish
with maturity stages III and IV in each 1 cm size class
(Rogers et al, 2009). The logistic curve was fitted by
minimising the sum of squares using the Solver ‘add-
in’ function in Excel (Microsoft Corporation, 2007).

Body condition and HSI analysis

Condition factor index, K and hepato somatic
index (HSI) were calculated as indirect indices of
energy status. The two parameters were estimated as
follows (McPherson et al. 2011):

a) HSI = 1 00 x LW (g) / S W (g)

b) K = 1 00 x S W (g) / FL 3 (cm)

RESULTS

Maturity stages

Catch rates for T. hamiltoni were much lower
than for M. pleurosticta , especially for males,
so maturity stages were pooled for each season
(three-month intervals). Both testes and ovaries of
M. pleurosticta showed a seasonal progression of
developmental stages, culminating in a switch from
a majority of fully mature fish in spring to a majority
of spent and resting individuals in summer (Fig. 2).
For T. hamiltoni the pattern is less clear but most
spent or resting individuals of both sexes were caught
in spring and summer whereas spawning-capable
fish were more common in autumn and winter. Fish
commencing a new gametogenic cycle were most
co mm on in summer (Fig. 2).

Gonadosomatic Index

Male and female M. pleurosticta both followed
similar patterns of monthly mean GSI (Fig. 3a). The
pattern was consistent for both years, elevated from
September and decreased in December. It is assumed
that the GSI value probably peaked in November but
no fish were captured in this month in either 2008
or 2009 presumably having left the mangrove habitat
to spawn elsewhere in the estuary or at sea. For I!
hamiltoni, GSI of females in both 2008 and 2009
showed a double peak with a maximum in April,
declining in May and increasing to a second larger
maximum in June or July (Fig. 3b). Catch rates for
males were too low to display meaningful patterns,
although they too peaked in April of both years.

Length at maturity

Fifty percent of female M. pleurosticta reached
sexual maturity at 89 mm and for males 50% reached
sexual maturity at 92 mm (Fig. 4a, b). At the total
length of 120 mm, all females were sexually mature
while all males above 130 mm were mature. In
contrast, male T. hamiltoni reached sexual maturity at
smaller size than females. Fifty percent of males were
sexually mature at 70 mm and by the total length at
80 mm all males were sexually mature whereas 50
percent of females attained sexual maturity at 83 mm
and all females were sexually mature at 1 10 mm (Fig.
4c, d).

Seasonal changes in GSI, K r and HSI

The ranges of seasonal mean condition factor
indices K were very similar for both sexes of both
species. For each species the seasonal patterns of
average HSI were similar for the two sexes and have
been pooled for analyses, but the patterns for the two
species are very different to each other (Fig. 5). The
seasonal mean hepatosomatic and gonadosomatic
indices have an inverse relationship to each other in
both species. In M. pleurosticta, there is little seasonal
change in K . However, both K and GSI values were
highest when HSI was decreasing. In T. hamiltoni, the
pattern of seasonal mean HSI is similar to K while
the GSI was peaking in winter when the HSI was at
its lowest.

DISCUSSION

Spawning season

Despite the superficial similarity of these two
species and their similar habitats, they display quite
different reproductive cycles. While M. pleurosticta
spawns in late spring, T. hamiltoni spawns in winter,
possibly with a split spawning in early and late

222

Proc. Linn. Soc. N.S.W., 136, 2014

R.B. MAT PIAH AND D.J. BUCHER

100%

80 %

60 %

40%

20 %

Summer Autumn Winter Spring
a. Females

100 %

80%

60 %

40%

20 %

Summer Autumn Winter
b, Males

Spring

Marilyna pleurosticta

100 %

80 %

60 %

40 %

20%

100 %

80%

60%

40%

20 %

0 %

Summer

Spring

Autumn Winter

c. Females Tetractenos hamiltoni

M Developing □ Spawning-capable

Summer Autumn Winter

d- Males
□ Spent/Resting

Spring

Fig. 2. Seasonal proportions of the three mature categories of gonad development for females (left) and
males (right) of M. pleurosticta (top) and I hamiltoni (bottom).

winter. The reproductive cycle of M pleurosticta in
this study was different to other tetraodontid species
reported, most of which spawn in summer (Habib
1979; Potter etal. 1988; Sabrahetal. 2006). However,
the spawning season for T. hamiltoni is very similar
to the closely related T. glaber in the Hawkesbury
estuarine system, which also spawns in winter (Booth
and Schultz 1999).

Spawning location

The consistent lack of captures of both species
in November and December of both years followed
by an increase in capture rates in subsequent months
suggests that both species probably leave the
mangrove systems at this time. Tag returns (Mat Piah
2011) suggest that at least some adults subsequently
return to the same channel system. Gonadosomatic
indices indicated that the majority of M. pleurosticta
caught prior to November were mature and the largest
proportion of stage IV (spent/resting) individuals

were captured in the months shortly after November,
indicating that the absence of that species coincided
with spawning. This finding is similar to the studies
in Swan River estuary (Potter et al. 1988), where
mature T. pieurogramma migrate out of the estuary
to spawn in shallow coastal waters between October
and January. At these times, large schools of this
species have often been observed passing out to sea
by fishers (Potter et al. 1988). T. hamiltoni were also
not caught in November and December of both years.
However, at this time their declining GSI suggests
that, while this species also leaves the mangroves in
early summer it is for reasons other than a spawning
migration.

Length at maturity

Although gonad development and subsequent
spawning may depend on various environmental
stimuli, individuals must reach a certain age or size
before they are capable of spawning (King 2001),

Proc. Linn. Soc. N.S.W., 136, 2014

223

Gonad osomatic Index Gonadosomatic Index

REPRODUCTIVE BIOLOGY OF ESTUARINE PUFFERFISH

Sampling Date

Fig. 3. Monthly mean (± S.E.) of Gonadosomatic Index (GSI) of females and males of (a) M. pleuros-
ticta and (b) T. hamiltoni.

224

Proc. Linn. Soc. N.S.W., 136, 2014

R.B. MAT PIAH AND D .J. BUCHER

& eo.o

100.0

8.0 10.0
Total Length (cm)

$. Females

12.0

14. D

Marilyna pleurosticta
1000

V Males

™ 80.0

v 60.0

I ‘>0.0

3 20 -°

0.0

4.0 8.0 3.0 10 0

Total Length (cm)

c, Females

14.0

0.0

2 0

4.0

34.0

Males

Tetractenos hamiltoni

Fig. 4. Length at maturity for M. pleurosticta (a and b) and T. hamiltoni (c and d). Lengths at which
50% of individuals are mature are indicated.

and dependency of maturation to the age or length is
strongly linked to growth and is also regulated by the
water temperature and feeding success (Yoneda et al.
2001; Takemura et al. 2004).

In this study, female M. pleurosticta
reached sexual maturity at a similar size to males,
whereas, male T. hamiltoni start to mature at a much
smaller size than females. It is not known if these
size differences are due to different growth rates or
different ages at maturity. Preliminary unvalidated
age estimates from otolith growth checks suggest that
age at maturity for both species may be as high as
10 years (Mat Piah 2011). The length at maturity in
pufferfish in this study was similar to that of U. richei
that matures at 7.5 to 11.6 cm (Habib 1979) while
Sabrah et al. (2006) determined that the large oceanic
species Lagocephalus sceleratus in the Gulf of Suez
reached maturity at a length of 42. 1 cm for males and
43.3 cm for females.

Relationship between energy storage and
reproductive activity

Many studies calculate K using total
weight including the gonads. In the case of puffers

that have relatively light bodies with reduced
skeletal components and large gonads and liver,
such formulae would have been strongly influenced
by individual organ development and variable gut
fullness, potentially masking changes in muscle and
fat body mass. For this reason, K was calculated
by excluding visceral weight from the numerator.
The cyclical variation in hepatosomatic index while
condition changed much less suggests a central role
for the liver a source of lipid and metabolic energy
fuelling gamete production.

There is usually a direct correlation between
hepatosomatic index and body condition index, and an
inverse correlation of these factors to gonadosomatic
index (Htun-han 1978). In this study HSI displayed a
strong inverse correlation with GSI in both species,
but K in M. pleurosticta showed no correlation
with either of the other two indices. However, in
T. hamiltoni K r showed a positive relationship with
HSI. The increase in GSI during the period of gonad
maturation is mainly due to the deposition of large
amounts of proteins and lipids in the developing
eggs and spermatozoa (Htun-han 1978). Part of
this material comes directly from ingested food

Proc. Linn. Soc. N.S.W., 136, 2014

225

REPRODUCTIVE BIOLOGY OF ESTUARINE PUFFERFISH

Summer Autumn Winter Spring

3.0

2.8

2.6

2.4

2.2

2.0

Summer

Autumn

Winter

■ Gonadosomatic index — Hepatosomatic Index

Spring

— — Condition Index

Fig. 5. Seasonal relationships between mean hepatosomatic Index (HSI), condition factor (Kr) and
gonadosomatic index (GSI) of (a) M. pleurosticta and (b) T. hamiltoni (sexes combined).

but a major proportion comes from reserves of
food deposited during the active feeding season in
organs such as liver and muscles (Larson 1974). It is
therefore reasonable to expect that the weight of liver
and muscle would reflect the cycle of accumulation
and utilization of these energy reserves.

Summer is important for both species because it
is the time of greatest growth after spawning as energy
is going into somatic tissue rather than gonads. The
increase in HSI also indicates an excess of energy
intake over immediate needs in this season. Thiswould

be a critical time for the species and feeding success at
this time could affect fecundity in the next spawning.
A flood in January 2008 resulting in low dissolved
oxygen and low pH for several months afterwards
may have substantially affected these, and potentially
other species during the first year of this study by
reducing feeding opportunities during this critical
period and may explain the interannual differences in
the proportions of mature and spent gonads, which
were more co mm on in samples from 2009 than in
2008.

226

Proc. Linn. Soc. N.S.W., 136, 2014

R.B. MAT PIAH AND D. J. BUCHER

The strong inverse relationship between HSI
and GSI suggests that lipid storage in the liver is
critical to reproductive success in both species. With
mean HSI values of between 6 and 10% of somatic
weight, the liver of these small pufferfish is unusually
large for the size of the fish. For comparison, other
fish species collected from the Richmond River in
August 2004 (Bucher, unpublished data) produced
mean HSI values for mature bream Acanthopagrus
australis of 0.7 percent, luderick Give/ la tricuspidata
of 1.0 percent and sand whiting Sillago ciliata of
1.8 percent. Mean GSI values of pufferfish at full
maturity (8-9) are also large compared to mature
bream Acanthopagrus australis of 4.0, luderick
Girella tricuspidata of 7.6 and sand whiting Sillago
ciliata of 2.1 (Bucher, unpublished data).

There was a different pattern in the relative
dynamics of HIS, K and GSI between these two
species. Marilyna pleurosticta starts reserving lipids
during the pre-spawning period, for use during
spawning. The same pattern was found in Irish Sea
plaice, Pleuronectes platessa (Wingfield and Grimm
1 977), where HSI was also highest in the pre-spawning
period and lowest in the post-spawning period. The
reduction in HSI and K over the reproductive season
may be explained by mobilization of lipid reserves,
and especially vitellogenin (Vg), a lipophosphoprotein
yolk-precursor synthesized by the liver (Maldonado-
Garcia et al. 2005). Somatic condition mirrored the
pattern of HSI in both species, indicating that lipids
stored outside the liver are also important in fuelling
the reproductive process.

In T. hamiltoni, HSI and K declined during
pre-spawning presumably for gonad maturation.
This pattern is similar to that of smooth pufferfish
Tetractenos glaber in the Sydney region (Booth and
Schultz 1999). In a study of liver weights of brook
trout Larson (1974), suggested that the decrease in
liver weight during pre-spawning season was due to
the passage of materials from the liver to the gonads
and concluded that weight changes of the liver plays
an important role in gonad maturation.

There was not only a change in weight but also a
change in colour and texture of the liver with different
stages of the gametogenic cycle. The pre-spawning
liver of both species was firm and pale while the
post-spawning liver was soft, dark and flaccid. This
supports the concept that that lipids in the liver have
been used for the spawning process. Rossouw (1987)
reported that the liver colour was in synchrony with
the variation in the total liver lipid content in both
sexes of sand sharks. He found that the higher liver
lipid concentration in the liver, the lighter the livers
become in appearance.

High values for hepatosomatic indices and their
strong inverse relationship with gonadosomatic
indices demonstrate that mobilisation of lipid
stores in the large liver is important for fuelling
gametogenesis and low feeding success during
periods of high river flow at critical times of the year
could therefore potentially severely affect spawning
success. The large variability in lipid content of the
liver has implications for tissue loads of lipophilic
pollutants if these ubiquitous, long-lived fishes are
to be used as biomonitors of estuarine pollution. The
differences in timing of lipid mobilisation for gonad
development also mean that the species being used for
a biomonitoring program will determine the timing of
sampling.

ACKNOWLEDGEMENTS

We gratefully acknowledge the technical assistance
by staff at the Biology Laboratory, School of Environment,
Science and Engineering, in particular Craig Taylor. The
project was funded by a Postgraduate Research Grant from
Southern Cross University and a Malaysian Government
Scholarship. Research was conducted under NSW DPI
permit no: P08/0031 and SCU An imal Care and Ethics
Approval 09/20.

REFERENCES

Allen, G. (2009). ‘Field Guide to Marine Fishes of
Tropical Australia and South-east Asia’. 4 th ed.
(Western Australian Museum, Perth).
Alonso-Femandez, A. and Saborido-Rey, F. (2012).
Relationship between energy allocation and
reproductive strategy in Trisopterus luscus. Journal
of Experimental Marine Biology and Ecology 416 ,
8-16.

Alquezar, R., Markich, S.J. and Booth, D.J. (2006). Metal
accumulation in the smooth toadfish, Tetractenos
glaber, in estuaries around Sydney, Australia.
Environmental Pollution, 142 , 123-131.

Baker, J.L. (2009). ‘Marine species of conservation
concern in South Australia. Vol.l-Bony and
cartilaginous fishes.’ South Australian Working
Group for Marine Species Conservation Concern.
Available at http://www.reefwatch.asn.au/pages/
bin/view/Publication/SamsccTETRAODONTIDAE
(Accessed 1 September 2010).

Bell, J.D., Pollard, D.A., Burchmore, J.J., Pease, B.C.
and Middleton, M.J. (1984). Structure of a fish
community in a temperate tidal mangrove creek in
Botany Bay, New South Wales. Australian Journal of
Marine and Freshwater Research 35 , 33-46.

Burton, M.P.M. (1994). A critical period for nutritional
control of early gametogenesis in female winter

Proc. Linn. Soc. N.S.W., 136, 2014

227

REPRODUCTIVE BIOLOGY OF ESTUARINE PUFFERFISH

flounder, Pleuronectes americanus (Pisces:

Teleostei). Journal of Zoology 233, 405-415.

Booth, D.J. and Schultz, D.L. (1999). Seasonal ecology,
condition and reproductive patterns of the smooth
toadfish, Tetractenos glaber (Freminville) in
the Hawkesburry estuarine system, Australia.
Proceedings of the Linnnean Society of New South
Wales 121, 71-84.

Brown-Peterson, N.J., Wyanski, D.M., Saborido-Rey,

F., Macewicz, B.J. and Lowerre-Barbieri (2011). A
standardized terminology for describing reproductive
development in fishes. Marine and Coastal Fisheries:
Dynamics, Management and Ecosystem Science 3,
52-70

Chellapa, S., Huntingford, F.A., Strang, R.H.C. and
Thomson, R.Y. (1995). Condition factor and
hepatosomatic index as estimates of energy status
in male three-spined stickleback. Journal of Fish
Biology? 47, 775-787.

Coates, D. (1993). Fish ecology and management of the
Sepik-Ramu, New Guinea, a large contemporary
tropical river basin. Environmental Biology ; of Fishes
38, 345-368.

Edgar, G.J. (2000). ‘Australian marine life: the plants and
animals of temperate waters’. (2 nd ed., Reed New
Holland: Australia).

Encina, L. and Granado-Lorencio, C. (1997). Seasonal
changes in condition, nutrition, gonad maturation and
energy content in barbel, Barbus sclateri, inhabiting
a fluctuating river. Environmental Biology of Fishes
50, 75-84

Fletcher, G.L. and King, M.J. (1978). Seasonal dynamics
of Cu 2+ , Zn 2+ , Ca 2+ , and Mg 2+ in gonads and liver of
winter flounder ( Pseudopleuronectes americanus ):
evidence for summer storage of Zn 2+ for winter
gonad development in females. Canadian Journal of
Zoology 56(2):284-290

Grant, E. (1987). ‘Fishes of Australia’. (E.M. Grant Pty.
Ltd: Redcliffe, Australia).

Gomon, M., Bray, D. and Kuiter, R. (2008). ‘Fishes of
Australia’s Southern Coast’. (Museum Victoria,
Melbourne).

Habib, G. (1979). Reproductive biology of the pufferfish,
Uranostoma richei (Plectognathi: Lagocepalidae),
from Lyttelton Harbour. New Zealand Journal of
Marine & Freshwater Research 13, 71-78.

Hindell, J. S. and Jenkins G. P. (2004). Spatial and

temporal variability in the assemblage structure of
fishes associated with mangroves ( Avicennia marina )
and intertidal mudflats in temperate Australian
embayments. Marine Biology; 144, 385-395.

Htun-han, M. (1978). The reproductive biology of the dab
Limanda limanda (L.) in the North Sea: gonosomatic
index, I H and condition factor. Journal of Fish
Biology 13, 369-378.

King, M. (2001). ‘Fisheries Biology, Assessment and
Management’. (Fishing News Books: Oxford,
England).

Koops, M.A., Hutchings, J.A. and McIntyre, T.M. (2004).
Testing hypotheses about fecundity, body size and

maternal condition in fishes. Fish and Fisheries 5,
120-130.

Larson, G.L. (1974). Liver weight of brook trout in a high-
mountain lake in Washington State. Progressive Fish
Culturalist 35, 234-236.

Maldonado-Garcia, M., Gracia-Lopez, V., Carillo, M.,
Hemadex-Herrera, A. and Rodriguez-Jaramillo,

C. (2005). Stages of gonad development during
the reproductive cycle of the blackfinsnook,
Centropomusmedius Gunther. Aquaculture Research
36, 554-563.

Mat Piah, R. (2011). Aspects of the ecology of small
estuarine pufferfish relevant to their value as
biomonitors of pollution. PhD thesis. Southern Cross
University, Lismore, NSW.

McPherson, L.R., Slotte, A., Kvamme, C., Meier, S.,
and Marshall, C.T. (2011). Inconsistencies in
measurement of fish condition: a comparison of four
indices of fat reserves for Atlantic herring ( Clupea
harengus). ICES Journal of Marine Science, 68,
52-60.

Merayo, C.R., (1996). Seasonal changes in the

biochemical composition of the muscle and liver of
bib ( Trisopterus luscus L.) (Pisces, Gadidae) from
the Cantabrian Sea (N Spain), Scientia Marina, 60,
489-495.

NSW Department of Primary Industries (2006). ‘A

strategic approach to the management of ornamental
fish in Australia’. Natural Resource Management
Ministerial Council. Available at http://www.dpi.

nsw.gov.au/ data/assets/pdffile/00 1 1/288425/

Management- of- omamental-fi sh-in- Australia .pdf
(Accessed 12 June 2009).

Potter, I.C., Cheal, A.J. & Loneragan, N.R. (1988).
Protracted estuarine phase in the life cycle of the
marine pufferfish, Torquigener pleuro gramma.
Marine Biology 98, 317-329.

Rogers, P.J., Ward, T.M., McLeay, L.J., Lowry, M.,

Saunders, R.J. and Williams, D. (2009). Reproductive
biology of blue mackerel, Scomber australasicus, off
southern and eastern Australia: suitability of the daily
egg production method for stock assessment. Marine
and Freshwater Research 60, 187-202.

Rossouw, G.J. ( 1987). Function of the liver and hepatic
lipids of the lesser sand shark, Rhinobatus annulatus
(Muller & Henle). Comparative Biochemistry
Physiology 86B, 785-790

Sabrah, M.M., El-Ganainy, A. A. and Zaky, M.A. (2006).
Biology and toxicity of the pufferfish Lagocephalus
sceleratus (Gmelin, 1789) from the Gulf of Suez.
Egyptian Journal of Aquatic Research 32, 283-297.

Stobutzki, I., Miller, M. and Brewer, D. (2001).
Sustainability of fishery bycatch: a process for
assessing highly diverse and numerous bycatch.
Environmental Conservation 28, 167-181.

Takemura A., Rahman M.S., Nakamura S., Park Y.J. and
Takano, K. (2004). Lunar cycles and reproductive
activity in reef fishes with particular attention to
rabbitfishes. Fish and Fisheries 5, 317-328.

228

Proc. Linn. Soc. N.S.W., 136, 2014

R.B. MAT PIAH AND D.J. BUCHER

Wingfield, J. C. and Grimm, A. S. (1977) Seasonal
changes in plasma cortisol, testosterone and
oestradiol-17(3 in the plaice, Pleuronectes platessa L.
General and Comparative Endocrinology 31 , 1-11
Yoneda, M., Muneharu T., Fujita H., Takeshita N.,
Takeshita K., Matsuyama M. and Matsuura S.

(2001). Reproductive cycle, fecundity, and seasonal
distribution of the anglerfish Lophius litulon in the
East China and Yellow Seas. Fisheries Bulletin 99,
356-370.

Proc. Linn. Soc. N.S.W., 136, 2014

229

230

The Effect of Disturbance Regime on
Darwinia glaucophylla (Myrtaceae) and its Habitat

Carmen Booyens 1 , Anita Chalmers 2 and Douglas Beckers 3

1 School of Science & Mathematics, Lake Macquarie Campus, Avondale College of Higher Education,

Cooranbong, 2265, NSW, Australia.

2 School of Environmental & Life Sciences, Ourimbah Campus, The University of Newcastle,

Ourimbah 2258, NSW, Australia.

3 National Parks and Wildlife Service, Gosford Office,

Gosford 2250, NSW, Australia.

Published on 29 December 2014 at http://escholarship.library.usyd.edu.au/joumals/index.php/LIN

Booyens, C., Chalmers, A. and Beckers, D. (2014). The effect of disturbance regime on Darwinia
glaucophylla (Myrtacease) and its habitat. Proceedings of the Linnean Society of New South Wales 136 ,
231-244.

The effect of disturbance regime (time since last fire or slashing) on the vulnerable plant species, Darwinia
glaucophylla, was assessed on the Central Coast of New South Wales, Australia. The abundance, growth
and flowering of D. glaucophylla adults and abundance and growth of seedlings was measured within sites
that had either been recently burnt (< 5 years), long unburnt (> 14 years) or regularly slashed (30 cm above
ground) along a utility easement. Our results showed that D. glaucophylla was most abundant at slashed
sites, followed by recently burnt sites; it was present but not abundant at unslashed sites that were burnt >

14 years ago. Seedlings were only found at one, recently burnt site. Disturbance regime had no significant
effect on the timing or density of flowering. Fmit collected from sites with different disturbance regimes did
not germinate after exposure to various combinations of heat, smoke -water and/or scarification. Recently
burnt sites contained plants producing a significantly greater number of viable fruits compared to those
from other disturbance regimes. Fire and slashing altered the habitat of D. glaucophylla in different ways.
Our findings suggest that slashing promotes favourable conditions for adults by creating a habitat with
higher light and less competition. However, it is not apparent whether these same conditions are favourable
for seedling recmitment.

Manuscript received 4 August 2014, accepted for publication 3 December 2014.

KEYWORDS: conservation management, fire, flowering, germination, slashing, threatened, utilities
easement

INTRODUCTION

Many Australian plant species are considered
disturbance-dependent, while others are sensitive to
significant disturbance (in which case the disturbance
may become a threatening process) (Ross et al. 2004;
Kirkpatrick 2007). To be a threat, the disturbance
must deleteriously interfere with transfers in the
life cycle of a species and/or significantly affect the
number of individuals at a particular life stage (Keith
1996). Fire is a natural disturbance that can pose
a threat to some species if the long-term regime is
disrupted in some way (Keith 1996). Fire frequency,
fire interval variability, fire intensity, season of burn
and pattern of bum are all elements of a fire regime

(Gill 1975; Bond and van Wilgen 1996) which, when
considered on a landscape scale, affect biodiversity
(Keith 1996). Keith (1996) identified twenty possible
fire-driven mechanisms of plant extinction. He
concluded that high and low fire frequency, as well
as repeated fires with little heat penetration of the
soil or the production of smoke derivatives, are fire
regimes likely to result in plant population decline
and extinction (Keith 1996). Therefore, management
of rare plants in fire-prone habitats typically requires
knowledge of life-cycle attributes critically involved
in population processes and the population response
to different fire regimes.

Other disturbances common to urban habitats
include sewer, water, gas and electricity services, all

EFFECT OF DISTURBANCE ON DARWIN I A GLAUCOPHYLLA

of which require installation of hard infrastructure,
often at the expense of biotic components of a
landscape (Foreman 2003). Slashing, whether by
hand or machinery, is the main means by which
utilities easements are maintained. Slashing allows
easy access for maintenance and surveillance, and
reduces fuel loads in order to decrease the threat of
fire on such services. Slashing of easements may
advantage some plant species, such as those well
represented in earlier successional stages and which
would subsequently be less well represented in mature
ecosystems. The current area of occupancy of D.
glaucophylla includes regularly slashed gas pipeline
and powerline easements located within National
Parks, raising the question as to whether slashing
is beneficial or detrimental to the species. Although
there has been a great deal of research on the effects
of slashing (hay-cropping) in grassland ecosystems
in Europe, where it is used as a management tool to
restore plant diversity to former agricultural land (see
review by Walker et al. 2004), only one study (Ellis
and Allen 2013) could be found on the impacts of
slashing in coastal heathland vegetation in Australia.

Darwinia glaucophylla B.G.Briggs is listed as
vulnerable under schedule 2 of the NSW Threatened
Species Conservation Act 1995. It is a prostrate shrub
found in fire -prone coastal heath where it occurs on
skeletal soils surrounding Hawkesbury sandstone
outcrops in the Gosford Local Government Area
(Department of Environment Climate Change and
Water 2009a). Its small extent of occurrence, high
endemism and habitat specificity has afforded this
vulnerable status (Department of Environment
Climate Change and Water 2009b). Previous studies
of D. glaucophylla include descriptive observations
of its morphology and phenology (Briggs 1962), seed
germination response to heat (Auld and Ooi 2009) and
the role of myrmecochory (Auld 2009). Auld and Ooi
(2009) found that heat (80°C) enhanced germination
and reported that seedlings emerge in the field 2-3
years after fire. However, the effect of smoke on the
seed germination of D. glaucophylla has not yet been
determined.

The current study aims to increase our
understanding of the ecology of D. glaucophylla in
a way that informs the management of the species.
As the species grows in fire-prone habitat and is
conspicuous in slashed areas along sections of the
Sydney to Newcastle gas/oil pipeline, but rarely
detected in adjacent unslashed areas, we ask the
following research questions: (1) Is the above ground
abundance of D. glaucophylla in slashed easements
and unslashed sites (adjacent to easements) similar
? (2) Does the above ground abundance of D.

glaucophylla differ between sites that have been
burnt in recent times compared with sites that were
burnt more than a decade ago? (3) Do the physical
characteristics of the habitat of D. glaucophylla differ
among disturbance regimes (fire, slashing) and, if
so, how? (4) Does the flowering phenology of D.
glaucophylla differ between disturbance regimes
(fire, slashing)? and (5) Does smoke water, heat and /
or scarification enhance seed germination?

METHODS

Study area

NSW National Parks and Wildlife Service
(NPWS) atlas records were used to choose four main
locations (Figure 1) within the extent of occurrence
of the species: Popran National Park (1 5 1 ° 1 3 ’05”E,
33°26 , 09”S), Girrakool Track (151°15’44”E,
33°25 , 46”S), Lyre Trig (151°17’51”E, 33°27 , 06”S)
and Rifle Range road (151 0 16’34”E, 33 0 27’24”S).
The latter three locations are all within Brisbane
Water National Park. The four locations were no more
than 12 kilometres apart and their elevation ranged
between 50 to 250 m ASL (Table 1). The Central
Coast region of NSW has a warm, temperate climate
and a summer maximum rainfall distribution (Murphy
1993). Mean annual rainfall at Narara Meteorological
Station (29 years of record) is 1280 mm (Bureau of
Meteorology 2009). This station was the closest to
most of the sites in this study. The mean maximum
temperature of 23° C occurs in January and the mean
minimum temperature of 1 1 ° C occurs in July (Narara
Meteorological Station 12 years of record) (Bureau
of Meteorology 2009). According to Murphy (1993),
the Rifle Range, Lyre Trig and Girrakool locations
belong to the Lambert soil landscape, having
undulating to rolling hills on Hawkesbury sandstone.
Slopes are typically < 20% and rock benches are
common (Murphy 1993). Soils are shallow and sandy
and within a pH 3.5 - pH 5.5 range (Murphy 1993).
Benson (1986) has categorised the vegetation at these
locations as consisting of open forest, woodland, open
scrub, open heath and sedgeland. Characteristic flora
present includes Banksia spp., Hakea spp., Grevillea
spp., Kunzea spp., Dillwynia spp., Acacia spp. and
Leptospermum spp. The Popran location differs in that
it belongs to the Gymean soil landscape, but it also has
a substrate comprising Hawkesbury sandstone with
similar vegetation communities to the other locations
(Benson 1986; Murphy 1993).

At each location, a plot measuring 10 m x 100
m (1000 m 2 ) was established in an area where D.
glaucophylla was present. At two of the locations

232

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

Brisbane
Water NP

Kariong

Brisbane
Water NP

Ourimbah

- ■

Gosford

Pacific

Ocean

Terrigal

0 2km

Figure 1. The four Darwinia glaucophylla sampling locations on the Central Coast of New
South Wales.

(Girrakool and Popran), adjacent unslashed and
slashed plots were set up, giving a total of six 1000
m 2 plots (hereafter referred to as sites). An orthogonal
design, to test for interactive effects between slashing
and fire, was not possible because there were no sites
available that were both recently (< 5 years ago) burnt
and slashed (Table 1). Thirty lm 2 quadrats were placed
randomly within each site. The size and number of
quadrats was based on a pilot study by Booyens (20 1 0)
which found that lm 2 quadrats showed less variance
in the percentage cover of D. glaucophylla than 25m 2
quadrats. Post-hoc power analysis demonstrated that
for a one-way ANOVA conducted on the percentage
cover of D. glaucophylla, a high level of power
(> 0.9) could be obtained with twenty-seven 1 m 2
quadrats (Booyens 2010). The pilot study also

demonstrated that density could not be used as a
measure of abundance because D. glaucophylla has
a prostrate growth form and can root at the nodes.
Percentage cover of the species was estimated using
the projected foliage photos of MacDonald et al.
(1990). Frequency of occurrence (mix of ramets and
genets) within the 30 quadrats at each site was also
determined. During the field component of the study
(i.e. spring 2008) new apical growth of 30 randomly
selected branchlets at each site was measured with
a ruler each fortnight. New growth was easily
recognised by its non-woody texture and pink/red
colour at the tips of branches. Flowering density was
estimated by counting the total number of flowers
within each quadrat at fortnightly increments over a 3
month period (10/8/08 - 6/11/08). The number of D.

Table 1. Characteristics of the locations where Darwinia glaucophylla was sampled in this study.

Location

Slope (°)

Elevation

(m)

Mean Fire
interval
(years)

Time since
last fire
(years)

Slashing

Girrakool NP

3.5

8.5 a

biannually with hand-held
brush cutters at 30 cm

Popran NP

4.5

120

9 b

above ground
biannually with hand-held
brush cutters at 30 cm

Lyre Trig

230

9.3 C

above ground
No slashed sites available

Rifle Range

180

9.5 d

No slashed sites available

a Burnt in 1977, 1980 & 1994; b Burnt in 1980 & 1989; c Burnt in 1969, 1977, 1987, 2000 & 2006; d Burnt
in 1965, 1969, 1989, 1994 & 2003.

Proc. Linn. Soc. N.S.W., 136, 2014

233

EFFECT OF DISTURBANCE ON DARWIN I A GLAUCOPHYLLA

glaucophylla seedlings located within each quadrat
was recorded and each seedling was marked and their
survival monitored for the duration of the project.
The sites and quadrats were permanently marked for
the duration of the project.

For each quadrat, the mean height and percentage
cover of surrounding vegetation was recorded.
Photosynthetically active radiation (PAR) was also
measured 1 m above the ground using a LI-190SA
quantum light photometer and expressed as pmol/
sec/m 2 . Soil samples (0.1 m deep) were collected
from a stratified random subset of the 1 m x 1 m
quadrats (n = 30). Soil pH was determined using a
1:5 dilution and a Hanna pH meter and electrode
(Rayment 1992). Electrical conductivity (EC) was
measured in a similar manner with a Hanna meter
and electrode (Rayment 1992). Soil moisture was
determined using the gravimetric method of Rayment
(1992). Total nitrogen (mg/kg), phosphorous (mg/kg)
and percentage organic matter tests were performed
by Sydney Analytical Laboratories using Australian
Standard (AS) methods. Samples were dried, split and
crushed to 150 microns prior to testing. Phosphorus
levels were determined using H 2 S0 4 digestion (APHA
4500BF), nitrogen by the APHA 4500B method and
organic matter by the AS method 1289.4. 1.1.

The indehiscent fruits (containing one ‘large’
seed) of D. glaucophylla were collected shortly
after the majority of flowering had occurred in late
November and early December 2008. Fruits were
collected with forceps from the ground at the base
of plants in the 1 m 2 quadrats and those fruits from
each site were pooled. A total of about 1800 fruits
were collected across the four sites and represented <
10% of what was available. Fruits were not collected
from unslashed plots as low numbers of fruits
meant that collection would have been ecologically
irresponsible. Fruits were stored in paper envelopes
in a cool, dry place until a germination experiment

could be conducted (about six months).

The treatments chosen for the germination
experiment (Figure 2) were based on previous studies
(Auld and Scott 1995; Kenny 2000; Cochrane et al.
2002; Tierney and Wardle 2005) which showed that
smoke water, heat and piercing the fruit coat enhanced
germination in other species, including other Darwinia
species. The fruit/seed coat of half of the collected
fruit was pierced with a fine needle to reduce any
impedance to germination imposed by the seed coat
(Cochrane et al. 2002). The fruit were placed on agar
(15g/L) plates to minimise desiccation and the need
for repeated watering during the experimental period.
Twenty-five fruit per plate were set up in duplicate for
each treatment and placed in a germination cabinet
(set at 12 hrs light/dark and 25°C / 15°C).

Fruits were soaked for four hours in a 0.1% (w/
v) solution of Thiram (a fungicide) or in a second
solution containing both Thiram and 2% commercial
smoke water (Regen 2000) according to the methods
of Tierney (2006). Where a heat treatment was
performed, fruits were heated in an equilibrated
glass Petri dish at 80° C for 10 mins and then the
appropriate solution (Thiram or Thiram plus smoke
water) was added (Baskin & Baskin, 1998; Tierney
& Wardle 2005). Fruits were placed equidistant
on the agar Petri dishes, sealed with Petri film and
placed one layer deep in the germination cabinet.
Germination was then monitored for a period of two
weeks (Baskin and Baskin 1998; ISTA 2003; Mt
Annan staff pers. comm. 2009). At the conclusion
of the experiment, the viability of the ungerminated
fruits was assessed using the ‘cut’ test (Baskin and
Baskin 1998; Cochrane et al. 2001; Ooi et al. 2005;
Mt Annan staff pers. comm., 2009).

Statistical analyses

Univariate two-factor analysis of variance
(ANOVA) was used to test for significant differences

(at the 0.05 level)
among means for
each of the variables
measured and to test
for any significant
interactions
(Tabachnick and Fidell
1996). The available
combinations of
disturbance in the
field meant that the
ANOVAs involved
both orthogonal and
nested designs. The
orthogonal design

Figure 2. Germination experiment design. T1 = buffered Thiram, no smoke water,
no heat; T2 = buffered Thiram, smoke water, no heat; T3 = buffered Thiram, no
smoke water, heat; T4 = buffered Thiram, smoke water, heat; T5 = distilled water.
Each treatment was performed in duplicate (Note: Macroplot = Site).

234

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

tested the effect of slashing or not slashing on sites
within the same fire regime (i.e. time since last fire >
14 years), while the nested design tested the effect of
time since last fire at unslashed sites. For both types
of ANOVAs, homogeneity of variance was tested
using Cochran’s test and normality was tested using
the Shipiro-Wilk test. Where necessary, the data were
transformed (arcsine or In) to improve homogeneity.
If transformation did not improve homogeneity,
ANOVA was conducted on untransformed data as
ANOVA is reportedly fairly robust to departures
from this assumption (Underwood 1997a). Post hoc
comparisons were made using Tukey -Krammer tests.
As the flowering data was not independent from one
sample time to the next, a repeated measures ANOVA
was conducted to test for differences in flowering
over time (Tabachnick and Fidell 1996) and post hoc
comparisons were made using Scheffe (Ho 2006).
Correlations between variables were tested using non-
parametric Spearman’s rho as a number of variables
were not normally distributed. Chi-squared (x 2 ) tests
were used to test for significant differences among
categorical data such as seed viability. All statistical
analyses were conducted with JMP (version 8), SPSS
(version 17) or GMAV (Underwood 1997b).

RESULTS

Effects of disturbance on abundance and growth

Darwinia glaucophylla was present in 59 of the
180 (33%) quadrats sampled in this study, with the
highest frequency occurring in the slashed/fire >14
years ago disturbance regime (Table 2). Percentage
cover ranged from 1% to 90%, with a mean
percentage cover of 5.2% (± 1.2) across all quadrats.
Mean percentage cover was significantly (p < 0.0001)
higher in the slashed/fire > 14 years ago disturbance
regime compared to the two other disturbance regimes
(Figure 3). There was no significant difference in
cover between unslashed/fire < 5 years ago locations
and the unslashed/fire > 14 years ago locations
(Figure 3 and Table 2). The effect of slashing on
percentage cover was consistent across the sites (i.e.
no interaction between slashing and site).

Only 5 (3%) of the 180 quadrats in this study
contained seedlings, with a total of 12 individuals
being recorded. All of these seedlings occurred at
the Lyre Trig site, which had an unslashed/fire < 5
years ago regime. Seven of the 12 seedlings survived
over the 12 -month monitoring period, with five
being killed by off-road vehicular damage. Those
remaining showed an average increase in height/
length of 1.3 cm (range 0.7 cm to 2 cm) over the 12-

slash e d/fire^: 1 4y rs ago unslashe d/f ire £ By rs age unslashe d/fire £1 4yrs age

Disturbance regime

Figure 3. Mean percentage (%) cover of Darwinia glaucophylla for each disturbance re-
gime (n = 60). Columns with the same letter are not significantly (p < 0.05) different. Bars
represent ± 1 standard error of the mean.

Proc. Linn. Soc. N.S.W., 136, 2014

235

EFFECT OF DISTURBANCE ON DARWIN I A GLAUCOPHYLLA

Table 2. Frequency of occurrence (%), mean percentage (%) cover and mean apical growth of Darwinia
glaucophylla at the six study sites (n = 30 at each site). Within rows, means with the same letter are not
significantly different from one another (at the 0.05 level). The effect of slashing (orthogonal design)
compares the Girrakool and Popran sites, while the effect of fire (nested design) compares the unslashed
sites at Girrakool and Popran with the last two columns in the table.

Girrakool NP
fire >14 years

Popran NP
fire >14 years

Lyre Trig
fire <5 years

Rifle Range
fire <5 years

unslashed

slashed

unslashed

slashed

unslashed

Frequency
(% of quadrats
at each site)

3.3

53.3

46.7

Mean (± SE)
percentage
cover (%)

0 a

13 (±4) b

l(±l) a

12 (±3) b

2 (±l) a

3 (±l) a

Mean (± SE)
apical growth
(cm/month)

0 a

5.5 (±0.3) b

0 a

4.0 (±0.2) c

4.5 (±0.4) b/c

3.7 (±0.2) c

month period and none produced flowers during this
period. By comparison, the mean apical growth rate
(in one month) of mature, established plants was 4.4
cm. Disturbance had a significant effect (p= 0.0007)
on mean apical growth and was largely attributable
to the absence of growth under the unslashed/fire>
14years ago regime (Table 2).

Flowering

Of the quadrats containing D. glaucophylla (59),
all but one contained individuals producing flowers.
At the height of flowering intensity in spring, around
9000 flowers were counted within a combined area
of 58 m 2 . Repeated measures ANOVA showed
no significant difference in the mean number of
flowers among disturbance regimes, nor a significant
interaction between sampling time and disturbance
regime. However, there was a significant difference
in mean number of flowers among sampling times
(Figure 4). The mean number of flowers significantly
increased with subsequent visits until flowering
peaked in September, after which it began to decline.
Post hoc pair-wise analysis showed no significant
difference in mean number of flowers between time
1 & 7 but flowering at these times were significantly
different from time 2 & 6 (which were similar to one
another) and from time 3, 4 & 5 (which were similar
to each other). There was a significant difference
in the mean number of flowers between sites under
the same fire regime, which was due to the absence
of flowers in the Girrakool unslashed site (data not
shown).

Seeds

No seeds germinated despite the various
treatments used. Microscope examination of fruits
collected from the ground beneath and adjacent to
established adults revealed that only 94/1600 (5.9%)
were filled. Cut tests showed that of those fruits
containing material, 21/94 (22%) had potentially
viable seed. Across the four sites from which fruits
were collected, greater than 90% of fruits were
empty. The percentage of fruits containing potentially
viable seed was significantly greater (x 2 =14.8; d.f. =
1; p = 0.01) in the unslashed/fire < 5years ago sites
compared to the slashed/fire > 14years ago sites. Of
the four sites able to be sampled, Popran had no viable
material within the collected fruits.

Effects of disturbance on the habitat of D.
glaucophylla

Mean percentage cover of associated vegetation
(lm above ground level) across the 180 quadrats
was 44%. Significant differences among the three
disturbance regimes were found for this variable, with
slashed/fire > 14 years ago areas having significantly
lower percentage cover of associated vegetation
compared to that of the other two disturbance regimes
(Table 3). For sites burnt > 14 years ago, mean
percentage cover of associated vegetation (1 m above
the ground) was significantly (p<0.01) lower in the
slashed compared to the unslashed areas. However
at Girrakool, slashing had no significant effect on
percentage cover of associated vegetation. Further,
the unslashed site at Girrakool had significantly (p <
0.05) less cover of associated vegetation compared

236

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

Figure 4. Flowering density of Darwinia glaucophylla over time for each disturbance re-
gime (A= slashed/fire>14years ago, n= 33; ■ = unslashed/fire<5years, n=22; • = unslashed/
fire>14 years ago, n= 2). Monitoring commenced (Time 1) on the 10/8/08 and ceased (Time
7) on the 6/11/08.

to the unslashed site at Popran even though the time
since last fire was similar ( > 14 years ago). There was
no significant effect of time since last fire (unslashed)
on the percentage cover of associated vegetation
(Table 3).

Mean maximum height of vegetation across
all quadrats (n=180) was 1.7 m (range 0.15 m to
10 m). Mean maximum vegetation height differed
significantly amongst the three disturbance regimes
(p<0.0001) with vegetation under an unslashed/fire >
14 years ago regime being significantly taller than the
other two regimes (Table 3). Within the slashed/fire >
14 years ago regime, the maximum vegetation height
at Girrakool was significantly greater than at Popran.
Within the unslashed sites, time since last fire had a
significant effect with maximum vegetation height
being lower when fire was < 5 years ago compared
to fire > 14 years ago (Table 3). Photosynthetic active
radiation (PAR) was significantly affected by time
since last fire (Table 3). Among the unslashed sites,
mean PAR was lower in areas burnt >14years ago
compared to those burnt < 5years ago (Table 3). The
effect of site was due to PAR being similar at the two

unslashed/fire < 5 years ago sites (Lyre Trig and Rifle
range), while within the unslashed/fire > 14 years
ago regime, Girrakool had significantly higher PAR
compared to Popran. PAR of the slashed sites was
similar to sites with an unslashed/ fire < 5years ago
regime (Table 3).

Of the suite of soil variables investigated, most
showed significant differences between disturbance
regimes and/or sites, but there were no significant
interactions between these two factors (Table 3). Time
since last fire had a significant effect on percentage
soil moisture (Table 3), with areas burnt < 5 years ago
having lower mean percentage soil moisture compared
to areas burnt > 14 years ago (within unslashed sites).
Soil pH, electrical conductivity, percentage organic
matter, total soil nitrogen and phosphorus were not
significantly affected by time since last fire (Table 3).
However, the effect of fire on soil nitrogen was close
to significance (p = 0.052). Slashing had a significant
effect on soil pH, electrical conductivity, total soil
nitrogen and percentage of organic matter in the soil
(Table 3). Mean soil pH was significantly higher
while EC, organic matter and total soil nitrogen was

Proc. Linn. Soc. N.S.W., 136, 2014

237

EFFECT OF DISTURBANCE ON DARWINIA GLAUCOPHYLLA

Table 3. ANOVA results (F-values) for percentage cover of associated vegetation, mean height of
vegetation, photosynthetically active radiation (PAR) and soil chemistry within quadrats. SI (slashing),
St (site), F (fire) . L - R: the effects of slashing (orthogonal design) are conveyed in columns 2, 3 & 4
of the table while the effects of fire (nested design) are shown in columns 5 & 6. *p<0.05, **p<0.01,

***p<0.001.

Variable

Six St

St(F)

d.f

Percentage
(%) cover of
associated

3.00

7.03**

5.31*

4.00*

1.85

vegetation

Mean height
(m) of
vegetation

0.03

106.38***

7.19**

1.50

41.88*

PAR

25.39**

134.23***

18.05***

20.78**

183.3***

pH of soil

0.97

10.98**

1.18

0.72

0.65

Electrical

conductivity
(pS) of soil

0.26

11.79**

5.49*

1.17

0.02

Percentage (%)
soil moisture

0.00

2.30

15.52**

0.0018

7.20*

Total soil

nitrogen (mg /
kg)

1.28

7.98*

2.00

3.16

1.01

Total soil

phosphorous

(mg/kg)

0.17

1.27

3.91

9.32*

0.07

significantly lower in the slashed areas compared
to the unslashed areas. Total soil phosphorus and
percentage soil moisture were not significantly
affected by slashing (Table 3).

Several environmental variables were
significantly correlated with quadrats containing
D. glaucophylla and with percentage cover of the
species (Table 4). Presence of, and/or percentage
cover of, D. glaucophylla was negatively correlated
with mean maximum height of vegetation, percentage
cover of associated vegetation, soil moisture content
and soil EC , and positively correlated with PAR and
soil pH. Average apical growth was positively and
negatively correlated with PAR and maximum height
of vegetation, respectively. Seedling presence was
positively corrected with soil N, P, percentage organic
matter and EC, while percentage of viable seeds was
positively correlated with the maximum height of
vegetation, percentage cover of associated vegetation
and soil N (Table 4). Although these correlations
were significant, some of the correlation co-efficients
were small indicating that the relationship between
some of the variables was weak and therefore should
be treated with caution.

DISCUSSION

Abundance and habitat

While 33% of quadrats surveyed in this study
contained D. glaucophylla, mean percentage cover
was only 5%, indicating that the above ground
abundance of the species is lower than initial field
observations suggested. The patchy nature of growth
in specific habitats such as rocky shelves often gives
the impression of local abundance (Booyens pers.
ob. 2007). Large spreading mats of D. glaucophylla
at slashed sites (along the pipeline easement) also
give the impression of abundance, but our results
show that the mats in this location are due to the
intentional management of the over-storey. Darwinia
glaucophylla was also found in unslashed areas where
fire had passed through more than 14 years ago, but it
was less frequent compared to sites burnt less than 5
years ago. This finding is consistent with the reported
decline of heathland sub-shrubs during long fire
intervals, as a result of density-dependent interactions
(Keith 1996). Species that are subordinate in stature
are particularly prone to competitive elimination (in
the absence of disturbance) but these competitive

238

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

Table 4. Significant correlations between measured environmental variables and attributes of D.
glaucophylla. PAR - photosynthetically active radiation; EC = electrical conductivity; N = total
nitrogen; P = total phosphorus; OM = organic matter. *p<0.05, **p<0.01, ***p<0.001.

Environmental Variable

D. glaucophylla attribute

Spearman p

p-value

PAR (pmol/sec/m)

% cover

0.3565

***

PAR (pmol/sec/m)

Average apical growth (cm)

0.6510

***

Max. height of vegetation (m)

% cover

-0.4482

***

Max. height of vegetation (m)

Average apical growth (cm)

-0.5597

***

Max. height of vegetation (m)

% viable seeds

0.6365

***

% cover associated vegetation

% cover

-0.4121

***

% cover associated vegetation

% viable seeds

0.2893

Presence in quadrat

0.4110

% cover

0.5578

EC (pS/cm)

% cover

-0.4112

EC (pS/cm)

quadrats containing seedlings

0.4218

% moisture (field)

Presence in quadrat

-0.3746

N (mg/kg)

% viable seeds

0.4802

N (mg/kg)

quadrats containing seedlings

0.4661

P (mg/kg)

quadrats containing seedlings

0.3941

% OM

quadrats containing seedlings

0.4081

interactions only affect the standing plant life stages
of populations (Keith 1996). Most plant species from
fire-prone communities are expected to have soil seed
banks (Auld et al. 2000). Depending on the longevity
of their dormant seeds, these species may persist in
the community long after standing plants have been
eliminated (Keith 1996). Although not examined
in this study, it is expected that D. glaucophylla is
also present in the soil seed bank at the long unburnt
sites. Darwinia species are known to have persistent
soil seed banks (Auld and Ooi 2009), which would
allow hidden (below ground) populations of D.
glaucophylla to persist during fire intervals typical of
the current study.

Both slashing and time since last fire had a
significant effect on the attributes and habitat of D.
glaucophylla (Table 5). Overall our results show
that there were more similarities between the sites
with differing times since last fire (unslashed) than
between sites with different slashing regimes (Table
5). That is, the presence of slashing had a greater
number and magnitude of effects than time since last
fire. However, slashed habitats did resemble areas

burnt less than five years ago in the following ways:
vegetation of lower stature, greater light penetration
and less soil moisture compared with those areas burnt
more than 14 years ago and not slashed. The effects
of slashing were not all negative though; slashing
resulted in the greatest frequency and percentage
cover of D. glaucophylla.

As soil variables are inter-related, it is difficult
to isolate the importance of individual soil factors to
the abundance of D. glaucophylla (especially because
correlation does not confer causality). However, the
results confirm that the soil characteristics of the
species habitat were typical of that found in heathland
vegetation on Hawkesbury sandstone (Murphy 1 993)
and that mature D. glaucophylla individuals can
tolerate a range of nitrogen levels (230 - 880mg/
kg). The species ability to tolerate low nitrogen may
be possible because of existing ectomycorrhizal
associations (Booyens 2010). Darwinia glaucophylla
was more likely to be present, and more abundant,
in quadrats with higher soil pH and lower soil
moisture (Table 5). The fact that phosphorus levels
were similar across the different disturbance regimes

Proc. Linn. Soc. N.S.W., 136, 2014

239

EFFECT OF DISTURBANCE ON DARWIN I A GLAUCOPHYLLA

Table 5. Summary of effects of disturbance regime (time
since last fire or slashing) on D.glaucophylla and its habitat.
* Due to presence of permanent creek through this site.

Slashed/fire
> 14 years ago

High frequency of quadrats containing D .glancophylla

High % cover of D, glancophylla

High apical growth

No seedlings

Similar flowering density

Low seed viability

No overstorey

Low stature vegetation

High PAR
Mean pH 5.1
Low EC

Medium soil moisture (field)

LowN

Similar P

Low % OM

Unslashed/fire
< 5 years ago

Low frequency of quadrats containing D. glancophylla

Low % cover of D. glaucophylla

High apical growth

Seedlings present

Similar flowering density

Higher seed viability

No overstorey

Low stature vegetation

High PAR
Mean pH 4.9
High EC

Low soil moisture (field)

HighN

Similar P

High % OM

Unslashed/fire
> 14 years ago

Low frequency of quadrats containing D. glaucophylla

Low % cover of D. glaucophylla

No apical growth

No seedlings

Similar flowering density

No seeds collected

Overstorey present

Tall stature vegetation

Low PAR

Mean pH 4.8

• High EC

• High soil moisture (field)

• High N

• Similar P

• High % OM

(Table 5) is not surprising, as most studies have
found that soil chemical properties return to pre-fire
conditions within a year (Raison 1979) and two years
had elapsed since the most recent fire in the current
study. Any nutrient pulse resulting from soil heating

and ash residues would have since been
taken up by the existing vegetation, been re-
immobilised by microbes or lost by leaching
(Raison 1979). Despite this, there was a weak
positive correlation between the presence of
D. glaucophylla seedlings in a quadrat and
total soil nitrogen and phosphorus levels. As
seeds of the species contain no endosperm
(Auld and Ooi 2009) the ash-bed effect, which
provides a temporary nutrient-rich substrate
allowing enhanced seedling growth (Hobbs
2002), may be particularly important.

Flowering

Slashing had no significant effect on
flowering, with mean density of flowers
and progression of flowering over time
similar across the different disturbance
regimes. Flowering fecundity and timing
may be affected by resource availability
such as adequate soil moisture (Craine
2005). However in our study, differences in
soil resources (i.e. moisture and nitrogen)
between the treatments (Table 5) appeared to
have little effect on flowering. Differences in
flowering response between individual sites
under the same disturbance regime (e.g.
unslashed Popran and unslashed Girrakool)
may be attributable to other site-specific
features such as aspect and degree of shading
from vegetation surrounding the pipeline
easement. Despite being surrounded by
vegetation, the Popran unslashed site showed
much higher levels of flowering compared
to the unslashed Girrakool site. This may
be explained by the maximum height of
vegetation at the latter site being greater
and site elevation being considerably lower;
both factors likely to increase the degree of
shading. In agreement with previous studies
(Briggs 1962; Myerscough 1998) we found
that peak flowering occurs in September. Peak
flowering came earliest to the Rifle Range
(fire < 5 years) site and latest to the Girrakool
slashed site, further indicating some site
specific differences. Fecundity could not be
ascertained as an unexpectedly large number
of flowers developed, preventing each
marked flower being followed over time. It
is recommended that future studies of flowering in
this species use the density estimates presented here
to determine a suitable quadrat size or number of
branchlets to sub-sample.

240

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

Seeds and seedlings

Whilst one sampling period is insufficient to
make conclusions about seedling recruitment in this
species, the only site containing seedlings was Lyre
Trig, which was burnt two years prior to the study.
Darwinia glaucophylla is an obligate-seeder with a
soil-stored seed bank and therefore fire is important for
the recruitment of this species (Auld and Ooi 2009).
Auld and Ooi (2009) found that the viability of fresh
seed collected from D. glaucophylla was high (85-
94%), irrespective of site and the year of collection.
Our seed viability results aren’t comparable because
it was tested at the end of the germination experiment,
by which time the seeds were at least 7 months old.
Greater than 90% of fruits collected in the current
study were empty at the end of the germination trial
(as determined by a cut test). Possible reasons for
empty fruits include: abscission of immature fruits due
to weather conditions, lack of resources, competition
with developing fruit for limited resources, genetic
abnormalities or lack of appropriate insect pollinators,
post-dispersal decomposition of fruits deposited on the
soil surface, pre- or post-dispersal insect predation of
fruit contents, or decomposition while on agar plates
despite addition of fungicidal agents (Stephenson
1981; Baskin and Baskin 1998). Myrmecochory has
also been reported for Darwinia spp. in south-eastern
Australia, with removal of abscised fruits being rapid
(within 4-5 days) (Auld 2009). As fruits in the current
study were collected post-dispersal, our findings raise
the question as to whether ants could be selectively
removing filled, and potentially viable, fruits. We also
found that twice as many fruits containing viable seed
were collected from the unslashed recently burnt sites
compared to the unslashed long unburnt sites. This
finding is worthy of further study.

For most of the south eastern Australian
Darwinia species, a large proportion (80-100%) of
the seed is dispersed in a dormant state (Auld and Ooi
2009). However for D. glaucophylla, the proportion
of fresh seed that were dormant was lower (40-75%)
and varied considerably between sites and slightly
between years (Auld and Ooi 2009). The findings
of the current study confer with Auld & Ooi (2009)
in that most of the annually produced seed were
dormant (100% in our study). However, Auld and
Ooi (2009) also found that heat elicited a germination
response in D. glaucophylla although the response
was quite variable. Three of the four seed crops (two
sites over two collection years) germinated at low
rates (20 - 40%) without heat treatment (i.e. controls),
one seed crop showed a temperature response after
being exposed to 60 - 110°C and two seed crops
showed a response to 80 - 100°C (Auld and Ooi

2009). The fact that no viable seeds in the current
study germinated (with or without treatment) cannot
be explained by differences in methods between the
two studies, with the exception of the seed collection
method. The different methods of seed collection
meant that post-dispersal environmental conditions
would have differed between the two studies, raising
the possibility that a short period (about 1 month) of
exposure to ground surface conditions in the current
study may have induced secondary dormancy. This
hypothesis is worthy of further study because Auld
et al. (2000) suggested that D. biflora may exhibit
seasonal secondary dormancy.

The maximum age of the seedlings at Lyre Trig
is two years and is therefore consistent with Auld and
Scott (1995) who found that seedlings of this species
emerge within 2-3 years after fire. Auld and Ooi (2009)
suggest that most seedlings don’t establish as a result
of over storey competition. The post-fire soil nutrient
status and above average rainfall during the current
study may have been favourable for the growth of
seedlings (Keith and Tozer 2012). Unfortunately few
seedlings were available to monitor in this project
and a more extensive search for seedlings across
the species range, followed by a longer monitoring
period, is needed to improve knowledge of seedling
recruitment and survival rates in different habitats.
Adults growing in sites that had either been slashed or
burnt less than 5 years ago showed a growth spurt well
after peak flowering, which may have been associated
with the higher than average rainfall experienced in
February 2009. It cannot be determined whether
the absence of apical growth of individuals in the
unslashed sites that were burnt > 14 years ago was due
to competition for resources or an artefact of species
abundance being so low that it was less likely that a
plant with apical growth was encountered. However,
it has been previously noted (Hobbs 2002) that a post-
fire environment in Australian heath is conducive to
high rates of growth at ground level.

Management implications

One of the disturbances to which D. glaucophylla
is currently indirectly exposed is slashing of
overstorey vegetation along the Sydney to Newcastle
gas/oil pipeline. Slashing or mowing under power
lines and within other utility easements serves to
reduce biomass in an area, limiting fuel for potential
fires and improving visibility of, and access to, the
easement. In cases where biomass is removed, soil
nutrient status and levels of organic matter may be
adversely affected (Walker 2004). Findings generally
vary as to whether nitrogen (N) or phosphorous (P)
decline and the extent of the decline (Walker et al.

Proc. Linn. Soc. N.S.W., 136, 2014

241

EFFECT OF DISTURBANCE ON DARWIN I A GLAUCOPHYLLA

2004). The current study found that slashed sites had
significantly higher soil pH and lower EC, nitrogen
and organic matter compared to unslashed sites (Table
5). It is proposed that removal of biomass reduced
inputs of organic matter into soil, which would have
also reduced available nitrate and ammonium ions,
contributing to lower electrical conductivity of soil
water. Disturbance of the soil profile during the initial
laying of the pipeline in 1978 (> 30 years ago) may
also explain the lower levels of nitrogen and organic
matter along the easement if the subsoil material was
brought to the surface and left exposed.

While current slashing practices (hand-held cutter
30 cm above ground) along the Sydney to Newcastle
gas/oil pipeline easement appear to promote the
survival and growth of existing mature individuals,
it is not known how long these practices have been in
place. Given that the species has a life expectancy of
around 20 - 30 years (Auld and Scott 1995), there has
been insufficient time to see the effects of slashing
over several generations. Further, it is not known
whether slashing can provide conditions necessary
for future recruitment. The current study found that
slashing did not affect flowering but we were unable
to determine whether fruit production was affected.
Although Auld and Scott (1995) demonstrated that 6-
month old seed of D. glaucophylla is viable, the long-
term persistence of seeds in the soil is not known. If
fire related cues are the only mechanism by which
seed dormancy is broken and the species doesn’t
produce any non-dormant seeds, then the population
along the pipeline at Popran may only be temporary.
Keith et al. (2002) reports that “most heath species
with persistent seed banks also produce a fraction of
non-dormant seeds” (p. 214) but Auld and Ooi (2009)
show that the proportion of D. glaucophylla seeds
that germinate without treatment is low. Increased
nitrogen levels in recently burnt areas often contribute
to more successful establishment of seedlings (Bell et
al. 1999). Thus even if seeds were to germinate in
the slashed sites in the absence of fire, the lower soil
nutrient levels together with a lack of a nutrient pulse
after fire may limit seedling establishment. The life-
span of D. glaucophylla is around 20-30 years (Auld
and Scott 1995) and the site has not experienced a
fire event for around 20 years. Auld and Scott (1995)
recommend a 5-10 year minimum interval between
fires for this species, but this may not be practicable
under power lines or above the pipeline. It is likely
that existing populations in Popran outside the unbumt
easement are too far away to allow natural dispersal
and recolonisation if the slashed population (including
the soil-stored seed bank) was to reach the end of its
life. If smoke alone can promote germination, this may

provide a management alternative for populations
where ecological burns cannot be conducted. Field
trials using aqueous smoke extracts, pelletised smoke
products or pile burns could then be conducted and
recruitment monitored.

CONCFUSION

The current study confirms that fire and
slashing both have positive effects on the above
ground abundance of D. glaucophylla. Our results
demonstrate that the above ground abundance of D.
glaucophylla was very low in long unburnt sites,
unless they have been slashed. The above ground
abundance of D. glaucophylla was greater at sites that
were recently burnt compared to those burnt more
than a decade ago, and seedlings were only found at
one site that was burnt < 5 years ago. Fire and slashing
affected the habitat of D. glaucophylla differently.
While both types of disturbance reduced the biomass
of the surrounding vegetation and increased light
penetration, slashing also resulted in lower levels of
soil nutrients and organic matter. Ideally, components
of the fire regime other than time since last fire should
be investigated to refine fire management strategies.
However given the species’ restricted distribution,
finding sites with suitable fire regimes may not be
possible. Flowering in D. glaucophylla peaked in
September and flowering density followed a similar
pattern over time, irrespective of the disturbance
regime. The results of the germination experiment
indicated that further study of the seed ecology of this
species is required.

The finding that the above ground abundance of
D. glaucophylla is higher at slashed sites and flowering
rates are unaffected by slashing goes some way to
support the conclusion of Monsted and McMillan
(2007) that current slashing practices along the
Sydney to Newcastle pipeline in habitats containing
D. glaucophylla are not adversely affecting existing
mature individuals. However, insufficient time
has elapsed to examine the effects of slashing over
several generations. Although the soil seed bank of D.
glaucophylla has the potential to allow the population
to persist after the above ground plants reach the end
of their life span, it was not apparent from the current
study whether slashing creates conditions that are
favourable for seedling recruitment.

ACKNOWFEDGEMENTS

We would like to thank K. Smith of National Parks and
Wildlife (Gosford) for assistance with Arc View to create

242

Proc. Linn. Soc. N.S.W., 136, 2014

C. BOO YENS, A. CHALMERS AND D. BECKERS

fire maps and all those who assisted with field work. We
also wish to thank D. Warman and K. O’Neil for assistance
with germination trials and statistics respectively. Finally,
the authors would like to thank the journal referees for
their constructive comments. This study was conducted
under a Scientific licence (S 12556) from the NSW Office
of Environment and Heritage.

REFERENCES:

Auld, T. D. (2009). Petals may act as a reward:

myrmecochory in shrubby Darwinia species of south-
eastern Australia. Austral Ecology 34 , 351-356.

Auld, T. D., Keith, D. D. and Bradstock, R. A. (2000).
Patterns in longevity of soil seedbanks in fire -prone
communities of south-eastern Australia. Australian
Journal of Botany 48 , 539-548.

Auld, T. D. and Ooi, M. K.J. (2009). Heat increases
germination of water-permeable seeds of obligate-
seeding Darwinia species (Myrtaceae). Plant Ecology
200 , 117-127.

Auld, T. D. and Scott, J. (1995). Conservation of
endangered plants in urban fire-prone habitats.
Species and Habitats Conference Nov. 13-16 Idaho.
Baskin, C. C. and Baskin, J. M. (1998). ‘ Seeds’.

(Academic Press: San Diego).

Bell, D. T., King, L. A.and Plummer, J. A (1999).

Ecophysiological effects of light quality and nitrate
on seed germination in species from Western
Australia. Australian Journal of Ecology 24 , 2-10.
Benson, D.H. (1986). Vegetation of the Gosford and Lake
Macquarie area. Cunninghamia 1(4), 467- 489.
Bond, W. J. and van Wilgen, B. (1996). ‘ Fire and Plants’.

(Chapman and Hall: London).

Briggs, B. G. (1962). The New South Wales Species of
Darwinia. Contributions from the N.S.W. National
Herbarium 3(3), 129-143.

Bureau of Meteorology (2009). www.bom. gov.au
[accessed 12/12/2009]

Cochrane, A., Kelly, A., Brown, K. and Cunneen S.

(2002), Relationships between seed germination
requirements and ecophysiological characteristics
aid the recovery of threatened native plant species
in Western Australia. Ecological Management
& Restoration 3(1), 47-59.

Craine, J. M. (2005). Reconciling plant strategy theories
of Grime and Tilman. Journal of Ecology 93, 1041-
1052.

Department of Environment, Climate Change and Water
(2009a) Darwinia glaucophylla profile (online) http://
www.environment.nsw.gov.au/threatenedSpeciesApp/
profile.aspx?id=1020 [accessed 3/9/09]

Department of Environment, Climate Change and Water
(2009b) Key Threatening Process (online) http://
www.environment.nsw.gov.au/threatenedspecies/kevt
hreateningprocesses . [accessed 3/9/09]

Ellis, M. and Allen T. (2013). A trial of slash and bum
management of Coast Tea-tree Leptospermum

laevigatum on Wonthaggi Heathland. The Victorian
Naturalist 130 , 212-218.

Foreman, R. T. T., Sperling, D., Bissonette, J. A.,

Clevenger, A. P., Cutshall, C. D., Dale, V.H. Fahrig,
L., France, R., Goldman, C. R, Heanue, K., Jones,

J. A., Swanson, F. J., Turrentine, T. and Winter, T .C.
(2003). ‘ Road Ecology’ . (Island Press: Washington).

Gill , A.M. (1975). Fire and the Australian flora: a review.
Australian Forestry 38, 4-25.

Ho, R. (2006). ‘ Handbook of univariate and multivariate
data analysis and interpretation with SPSS’

(Chapman and Hall: London).

Hobbs, R. (2002). Fire regimes and their effects in
Australian temperate woodland. In ‘Flammable
Australia: The fire regimes and biodiversity of
a continent’ (Eds R.A. Bradstock, J.E. Williams
and M.A. Gill) (Cambridge University Press:
Cambridge).

International Seed Testing Association (ISTA) (2003).
International Rules for Seed Testing. Switzerland.

JMP (v 8.0) (2008). SAS Institute Inc, Cary, USA.

Keith, D.A. (1996). Fire-driven extinction of plant
populations: a synthesis of theory and review of
evident from Australian vegetation. Proceedings of
the Linnean Society NSW 116 , 37-78.

Keith, D.A., McCaw, W.L. and Whelan, R.J. (2002). Fire
regimes in Australian heathlands and their effects on
plants and animals. In ‘ Flammable Australia: The fire
regimes and biodiversity of a continent’ (Eds R.A.
Bradstock, J.E. Williams and M.A. Gill) (Cambridge
University Press: Cambridge).

Keith, D.A and Tozer, M.G. (2012). The influence of fire ,
herbivores and rainfall on vegetation dynamics in the
mallee: a long-term experiment. Proceedings of the
Linnean Society of NSW 134 , 40-54

Kenny, B. J. (2000). Influence of multiple fire -related
germination cues on three Sydney Grevillea
(Proteaceae) species. Austral Ecology 25, 664-669.

Kirkpatrick, J.B. (2007). Collateral benefit: unconscious
conservation of threatened plant species. Australian
Journal of Botany 55, 221-224.

MacDonald, R.C., Isbell, R.F., Speight, J.G., Walker, J.
and Hopkins, M.S. (1990). ‘Australian Soil and Land
Survey’ (2 nd ed). (Inkata Press: Melbourne).

Monsted, P. and McMillan, A . (2007). Sydney to
Newcastle Gas Pipeline Easement Biodiversity
Assessment and EMP for maintenance works. Parsons
Brinckerhoff, Sydney.

Muiphy, C.L. (1993). Soil Landscapes of the Gosford-

Lake Macquarie 1:100 000 Sheet Report, Department
of Conservation and Land Management.

Myerscough, P.J. (1998). Ecology of Myrtaceae
with special reference to the Sydney region.
Cunninghamia 5(4), 787-807.

Ooi, M., Auld, T. and Whelan, R. (2005). Comparison
of the cut and tetrazolium tests for assessing seed
viability : a study using Australian native Leucopogon
species. Ecological Management and Restoration
5(2), 141-143.

Proc. Linn. Soc. N.S.W., 136, 2014

243

EFFECT OF DISTURBANCE ON DARWINIA GLAUCOPHYLLA

Raison, R. J. (1979). Modification of the soil environment
by vegetation fires, with particular reference to
nitrogen transformations: a review. Plant and Soil 51 ,
73-108.

Rayment, G.E. (1992). ‘Australian laboratory handbook
of soil and water chemical methods (Inkata Press:
North Ryde).

Ross, K. A., Taylor, J. E., Fox, M. D. and Fox, B.

J. (2004). Interaction of multiple disturbances
importance of disturbance interval in the effects of
fire on rehabilitating mined areas. Austral Ecology
29 , 508-529.

Stephenson, A. G. (1981). Flower and fruit abortion:
proximate causes and ultimate functions. Annual
Review of Ecology & Systematics 12 , 253-279.

Tabachnick, B. G. and Fidell, F. S. (1996). ‘Using
Multivariate Statistics’ (3 rd ed). (Harper Collins
College Publishers: California).

Tierney, D. A. (2006). The effect of fire -related

germination cues on the germination of a declining
forest understory species. Australian Journal of
Botany 54 , 297-303.

Tierney, D.A. and Wardle, G. A. (2005). Differential seed
ecology in in the shrubs Kunzea rupestris, K. capitata
and associated hybrids (Myrtaceae): the function of
thin-walled fruit in a fire -prone vegetation. Australian
Journal of Botany 53, 313-321.

Underwood, A. J (1997a). ‘Experiments in Ecology: Their
Eogical Design and Interpretation Using Analysis of
Variance’. (Cambridge University Press: Cambridge).

Underwood, A. J. (1997b). ‘GMAV statistical software’.
(Cambridge University Press: Cambridge).

Walker, K.J., Stevens, P.A., Stevens, D.P., Mountford,

J.O., Manchester, S.J. and Pywell, R.F. (2004). The
restoration and re-creation of species-rich lowland
grassland on land formerly managed for
intensive agriculture in the UK. Biological
Conservation 119 , 1-18.

Proc. Linn. Soc. N.S.W., 136, 2014

244

The Linnean Society of New South Wales publishes in its proceedings original papers and review
article dealing with biological and earth sciences. Intending authors should contact the Secretary
( PO Box 82, Kingsford, N.S.W. 2032, Australia ) for instructions for the preparation of manuscripts
and procedures for submission. Instructions to authors are also available on the society’s web page
(http : / /linne ans o cietyns w. org.au/ ) .

Manuscripts not prepared in accordance with the society’s instructions will not be considered.

PROCEEDINGS OF THE LINNEAN SOCIETY OF NSW
VOLUME 136

f §/' THE
/§/ LINNEAN'
H SOCIETY

fe\ OF

NEW SOUTH
WALES

CONTENTS

Volume 136

Papers published in 2014, compiled 31 December 2014

Published at http://escholarship.library.usyd.edu.au/journals/index.php/LIN
(date individual papers were published online at eScholarship)

Volume 136

Compiled 31 December 2015
TABLE OF CONTENTS

Section 1 - Papers arising from a symposium held at Jenolan Caves in May 2013.

1-18 Pogson, R.E., Osborne, R.A.L. and Colchester, D.M.

Minerals of Jenolan Caves, New South Wales, Australia: geological and biological interactions.

19-34 Baker, A.W.

The Jenolan environmental monitoring program.

35-67 Eberhard, S.M., Smith, G.B., Gibian, M.M., Smith, H.M. and Gray, M.R.

Invertebrate cave fauna of Jenolan.

69-75 Bellamy, K. and Barnes, C.

Jenolan show caves: origin of cave and feature names.

77-97 Osborne, R.A.L.

Understanding the origin and evolution of Jenolan Caves: the next steps.

99-130 Branagan, D.F., Pickett, J.W. and Percival, I.G.

Geology and Geomorphology of Jenolan Caves and the Surrounding Region.

Section 2 - General papers

131-155 Martin, H.A.

A review of the Cenozoic palynostratigraphy of the River Valleys in Central and Western New South Wales.
157-199 Lunney, D.

Integrating history and ecological thinking: Royal National Park in historical perspective.

201-218 McLean, G.

A comparative study of the Australian fossil shark egg-case Palaeoxyris duni, with comments on affinities and
structure.

219-229 Mat Piah, R.B. and Bucher, D.J.

Reproductive biology of estuarine pufferfish, Marilyna pleurosticta and Tetractenos hamiltoni (Teleostei:
Tetraodontidae) in northern New South Wales: implications for biomonitoring.

231-244 Booyens, C., Chalmers, A. and Beckers, D.

The effect of disturbance regime on Darwinia glaucophylla (Myrtacease) and its habitat.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Proceedings of the Linnean Society of New South Wales"

See other formats